Add files

.gitattributes

@@ -25,3 +25,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zstandard filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text

LICENSE

@@ -0,0 +1,395 @@
+Copyright (c) 2022, National Diet Library, Japan.
+
+Attribution 4.0 International
+
+=======================================================================
+
+Creative Commons Corporation ("Creative Commons") is not a law firm and
+does not provide legal services or legal advice. Distribution of
+Creative Commons public licenses does not create a lawyer-client or
+other relationship. Creative Commons makes its licenses and related
+information available on an "as-is" basis. Creative Commons gives no
+warranties regarding its licenses, any material licensed under their
+terms and conditions, or any related information. Creative Commons
+disclaims all liability for damages resulting from their use to the
+fullest extent possible.
+
+Using Creative Commons Public Licenses
+
+Creative Commons public licenses provide a standard set of terms and
+conditions that creators and other rights holders may use to share
+original works of authorship and other material subject to copyright
+and certain other rights specified in the public license below. The
+following considerations are for informational purposes only, are not
+exhaustive, and do not form part of our licenses.
+
+     Considerations for licensors: Our public licenses are
+     intended for use by those authorized to give the public
+     permission to use material in ways otherwise restricted by
+     copyright and certain other rights. Our licenses are
+     irrevocable. Licensors should read and understand the terms
+     and conditions of the license they choose before applying it.
+     Licensors should also secure all rights necessary before
+     applying our licenses so that the public can reuse the
+     material as expected. Licensors should clearly mark any
+     material not subject to the license. This includes other CC-
+     licensed material, or material used under an exception or
+     limitation to copyright. More considerations for licensors:
+	wiki.creativecommons.org/Considerations_for_licensors
+
+     Considerations for the public: By using one of our public
+     licenses, a licensor grants the public permission to use the
+     licensed material under specified terms and conditions. If
+     the licensor's permission is not necessary for any reason--for
+     example, because of any applicable exception or limitation to
+     copyright--then that use is not regulated by the license. Our
+     licenses grant only permissions under copyright and certain
+     other rights that a licensor has authority to grant. Use of
+     the licensed material may still be restricted for other
+     reasons, including because others have copyright or other
+     rights in the material. A licensor may make special requests,
+     such as asking that all changes be marked or described.
+     Although not required by our licenses, you are encouraged to
+     respect those requests where reasonable. More_considerations
+     for the public: 
+	wiki.creativecommons.org/Considerations_for_licensees
+
+=======================================================================
+
+Creative Commons Attribution 4.0 International Public License
+
+By exercising the Licensed Rights (defined below), You accept and agree
+to be bound by the terms and conditions of this Creative Commons
+Attribution 4.0 International Public License ("Public License"). To the
+extent this Public License may be interpreted as a contract, You are
+granted the Licensed Rights in consideration of Your acceptance of
+these terms and conditions, and the Licensor grants You such rights in
+consideration of benefits the Licensor receives from making the
+Licensed Material available under these terms and conditions.
+
+
+Section 1 -- Definitions.
+
+  a. Adapted Material means material subject to Copyright and Similar
+     Rights that is derived from or based upon the Licensed Material
+     and in which the Licensed Material is translated, altered,
+     arranged, transformed, or otherwise modified in a manner requiring
+     permission under the Copyright and Similar Rights held by the
+     Licensor. For purposes of this Public License, where the Licensed
+     Material is a musical work, performance, or sound recording,
+     Adapted Material is always produced where the Licensed Material is
+     synched in timed relation with a moving image.
+
+  b. Adapter's License means the license You apply to Your Copyright
+     and Similar Rights in Your contributions to Adapted Material in
+     accordance with the terms and conditions of this Public License.
+
+  c. Copyright and Similar Rights means copyright and/or similar rights
+     closely related to copyright including, without limitation,
+     performance, broadcast, sound recording, and Sui Generis Database
+     Rights, without regard to how the rights are labeled or
+     categorized. For purposes of this Public License, the rights
+     specified in Section 2(b)(1)-(2) are not Copyright and Similar
+     Rights.
+
+  d. Effective Technological Measures means those measures that, in the
+     absence of proper authority, may not be circumvented under laws
+     fulfilling obligations under Article 11 of the WIPO Copyright
+     Treaty adopted on December 20, 1996, and/or similar international
+     agreements.
+
+  e. Exceptions and Limitations means fair use, fair dealing, and/or
+     any other exception or limitation to Copyright and Similar Rights
+     that applies to Your use of the Licensed Material.
+
+  f. Licensed Material means the artistic or literary work, database,
+     or other material to which the Licensor applied this Public
+     License.
+
+  g. Licensed Rights means the rights granted to You subject to the
+     terms and conditions of this Public License, which are limited to
+     all Copyright and Similar Rights that apply to Your use of the
+     Licensed Material and that the Licensor has authority to license.
+
+  h. Licensor means the individual(s) or entity(ies) granting rights
+     under this Public License.
+
+  i. Share means to provide material to the public by any means or
+     process that requires permission under the Licensed Rights, such
+     as reproduction, public display, public performance, distribution,
+     dissemination, communication, or importation, and to make material
+     available to the public including in ways that members of the
+     public may access the material from a place and at a time
+     individually chosen by them.
+
+  j. Sui Generis Database Rights means rights other than copyright
+     resulting from Directive 96/9/EC of the European Parliament and of
+     the Council of 11 March 1996 on the legal protection of databases,
+     as amended and/or succeeded, as well as other essentially
+     equivalent rights anywhere in the world.
+
+  k. You means the individual or entity exercising the Licensed Rights
+     under this Public License. Your has a corresponding meaning.
+
+
+Section 2 -- Scope.
+
+  a. License grant.
+
+       1. Subject to the terms and conditions of this Public License,
+          the Licensor hereby grants You a worldwide, royalty-free,
+          non-sublicensable, non-exclusive, irrevocable license to
+          exercise the Licensed Rights in the Licensed Material to:
+
+            a. reproduce and Share the Licensed Material, in whole or
+               in part; and
+
+            b. produce, reproduce, and Share Adapted Material.
+
+       2. Exceptions and Limitations. For the avoidance of doubt, where
+          Exceptions and Limitations apply to Your use, this Public
+          License does not apply, and You do not need to comply with
+          its terms and conditions.
+
+       3. Term. The term of this Public License is specified in Section
+          6(a).
+
+       4. Media and formats; technical modifications allowed. The
+          Licensor authorizes You to exercise the Licensed Rights in
+          all media and formats whether now known or hereafter created,
+          and to make technical modifications necessary to do so. The
+          Licensor waives and/or agrees not to assert any right or
+          authority to forbid You from making technical modifications
+          necessary to exercise the Licensed Rights, including
+          technical modifications necessary to circumvent Effective
+          Technological Measures. For purposes of this Public License,
+          simply making modifications authorized by this Section 2(a)
+          (4) never produces Adapted Material.
+
+       5. Downstream recipients.
+
+            a. Offer from the Licensor -- Licensed Material. Every
+               recipient of the Licensed Material automatically
+               receives an offer from the Licensor to exercise the
+               Licensed Rights under the terms and conditions of this
+               Public License.
+
+            b. No downstream restrictions. You may not offer or impose
+               any additional or different terms or conditions on, or
+               apply any Effective Technological Measures to, the
+               Licensed Material if doing so restricts exercise of the
+               Licensed Rights by any recipient of the Licensed
+               Material.
+
+       6. No endorsement. Nothing in this Public License constitutes or
+          may be construed as permission to assert or imply that You
+          are, or that Your use of the Licensed Material is, connected
+          with, or sponsored, endorsed, or granted official status by,
+          the Licensor or others designated to receive attribution as
+          provided in Section 3(a)(1)(A)(i).
+
+  b. Other rights.
+
+       1. Moral rights, such as the right of integrity, are not
+          licensed under this Public License, nor are publicity,
+          privacy, and/or other similar personality rights; however, to
+          the extent possible, the Licensor waives and/or agrees not to
+          assert any such rights held by the Licensor to the limited
+          extent necessary to allow You to exercise the Licensed
+          Rights, but not otherwise.
+
+       2. Patent and trademark rights are not licensed under this
+          Public License.
+
+       3. To the extent possible, the Licensor waives any right to
+          collect royalties from You for the exercise of the Licensed
+          Rights, whether directly or through a collecting society
+          under any voluntary or waivable statutory or compulsory
+          licensing scheme. In all other cases the Licensor expressly
+          reserves any right to collect such royalties.
+
+
+Section 3 -- License Conditions.
+
+Your exercise of the Licensed Rights is expressly made subject to the
+following conditions.
+
+  a. Attribution.
+
+       1. If You Share the Licensed Material (including in modified
+          form), You must:
+
+            a. retain the following if it is supplied by the Licensor
+               with the Licensed Material:
+
+                 i. identification of the creator(s) of the Licensed
+                    Material and any others designated to receive
+                    attribution, in any reasonable manner requested by
+                    the Licensor (including by pseudonym if
+                    designated);
+
+                ii. a copyright notice;
+
+               iii. a notice that refers to this Public License;
+
+                iv. a notice that refers to the disclaimer of
+                    warranties;
+
+                 v. a URI or hyperlink to the Licensed Material to the
+                    extent reasonably practicable;
+
+            b. indicate if You modified the Licensed Material and
+               retain an indication of any previous modifications; and
+
+            c. indicate the Licensed Material is licensed under this
+               Public License, and include the text of, or the URI or
+               hyperlink to, this Public License.
+
+       2. You may satisfy the conditions in Section 3(a)(1) in any
+          reasonable manner based on the medium, means, and context in
+          which You Share the Licensed Material. For example, it may be
+          reasonable to satisfy the conditions by providing a URI or
+          hyperlink to a resource that includes the required
+          information.
+
+       3. If requested by the Licensor, You must remove any of the
+          information required by Section 3(a)(1)(A) to the extent
+          reasonably practicable.
+
+       4. If You Share Adapted Material You produce, the Adapter's
+          License You apply must not prevent recipients of the Adapted
+          Material from complying with this Public License.
+
+
+Section 4 -- Sui Generis Database Rights.
+
+Where the Licensed Rights include Sui Generis Database Rights that
+apply to Your use of the Licensed Material:
+
+  a. for the avoidance of doubt, Section 2(a)(1) grants You the right
+     to extract, reuse, reproduce, and Share all or a substantial
+     portion of the contents of the database;
+
+  b. if You include all or a substantial portion of the database
+     contents in a database in which You have Sui Generis Database
+     Rights, then the database in which You have Sui Generis Database
+     Rights (but not its individual contents) is Adapted Material; and
+
+  c. You must comply with the conditions in Section 3(a) if You Share
+     all or a substantial portion of the contents of the database.
+
+For the avoidance of doubt, this Section 4 supplements and does not
+replace Your obligations under this Public License where the Licensed
+Rights include other Copyright and Similar Rights.
+
+
+Section 5 -- Disclaimer of Warranties and Limitation of Liability.
+
+  a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
+     EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
+     AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
+     ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
+     IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
+     WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
+     PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
+     ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
+     KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
+     ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
+
+  b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
+     TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
+     NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
+     INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
+     COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
+     USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
+     ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
+     DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
+     IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
+
+  c. The disclaimer of warranties and limitation of liability provided
+     above shall be interpreted in a manner that, to the extent
+     possible, most closely approximates an absolute disclaimer and
+     waiver of all liability.
+
+
+Section 6 -- Term and Termination.
+
+  a. This Public License applies for the term of the Copyright and
+     Similar Rights licensed here. However, if You fail to comply with
+     this Public License, then Your rights under this Public License
+     terminate automatically.
+
+  b. Where Your right to use the Licensed Material has terminated under
+     Section 6(a), it reinstates:
+
+       1. automatically as of the date the violation is cured, provided
+          it is cured within 30 days of Your discovery of the
+          violation; or
+
+       2. upon express reinstatement by the Licensor.
+
+     For the avoidance of doubt, this Section 6(b) does not affect any
+     right the Licensor may have to seek remedies for Your violations
+     of this Public License.
+
+  c. For the avoidance of doubt, the Licensor may also offer the
+     Licensed Material under separate terms or conditions or stop
+     distributing the Licensed Material at any time; however, doing so
+     will not terminate this Public License.
+
+  d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
+     License.
+
+
+Section 7 -- Other Terms and Conditions.
+
+  a. The Licensor shall not be bound by any additional or different
+     terms or conditions communicated by You unless expressly agreed.
+
+  b. Any arrangements, understandings, or agreements regarding the
+     Licensed Material not stated herein are separate from and
+     independent of the terms and conditions of this Public License.
+
+
+Section 8 -- Interpretation.
+
+  a. For the avoidance of doubt, this Public License does not, and
+     shall not be interpreted to, reduce, limit, restrict, or impose
+     conditions on any use of the Licensed Material that could lawfully
+     be made without permission under this Public License.
+
+  b. To the extent possible, if any provision of this Public License is
+     deemed unenforceable, it shall be automatically reformed to the
+     minimum extent necessary to make it enforceable. If the provision
+     cannot be reformed, it shall be severed from this Public License
+     without affecting the enforceability of the remaining terms and
+     conditions.
+
+  c. No term or condition of this Public License will be waived and no
+     failure to comply consented to unless expressly agreed to by the
+     Licensor.
+
+  d. Nothing in this Public License constitutes or may be interpreted
+     as a limitation upon, or waiver of, any privileges and immunities
+     that apply to the Licensor or You, including from the legal
+     processes of any jurisdiction or authority.
+
+
+=======================================================================
+
+Creative Commons is not a party to its public licenses.
+Notwithstanding, Creative Commons may elect to apply one of its public
+licenses to material it publishes and in those instances will be
+considered the "Licensor." Except for the limited purpose of indicating
+that material is shared under a Creative Commons public license or as
+otherwise permitted by the Creative Commons policies published at
+creativecommons.org/policies, Creative Commons does not authorize the
+use of the trademark "Creative Commons" or any other trademark or logo
+of Creative Commons without its prior written consent including,
+without limitation, in connection with any unauthorized modifications
+to any of its public licenses or any other arrangements,
+understandings, or agreements concerning use of licensed material. For
+the avoidance of doubt, this paragraph does not form part of the public
+licenses.
+
+Creative Commons may be contacted at creativecommons.org.

app.py

@@ -0,0 +1,40 @@
+import os
+import torch
+
+print(torch.__version__)
+torch_ver, cuda_ver = torch.__version__.split('+')
+os.system(f'pip install mmcv-full==1.4.0 -f https://download.openmmlab.com/mmcv/dist/{cuda_ver}/torch{torch_ver}/index.html --no-cache-dir')
+os.system('cd src/ndl_layout/mmdetection && python setup.py bdist_wheel && pip install dist/*.whl')
+os.system('wget https://lab.ndl.go.jp/dataset/ndlocr/text_recognition/mojilist_NDL.txt -P ./src/text_recognition/models')
+os.system('wget https://lab.ndl.go.jp/dataset/ndlocr/text_recognition/ndlenfixed64-mj0-synth1.pth -P ./src/text_recognition/models')
+os.system('wget https://lab.ndl.go.jp/dataset/ndlocr/ndl_layout/ndl_layout_config.py -P ./src/ndl_layout/models')
+os.system('wget https://lab.ndl.go.jp/dataset/ndlocr/ndl_layout/epoch_140_all_eql_bt.pth -P ./src/ndl_layout/models')
+os.system('wget https://lab.ndl.go.jp/dataset/ndlocr/separate_pages_ssd/weights.hdf5 -P ./src/separate_pages_ssd/ssd_tools')
+os.system("wget https://i.imgur.com/fSL1CGG.jpg")
+os.environ["PYTHONPATH"]=os.environ["PYTHONPATH"]+":"+f"{os.getcwd()}/src/text_recognition/deep-text-recognition-benchmark"
+
+import gradio as gr
+from uuid import uuid4
+from pathlib import Path
+
+
+def inference(im):
+    dir_name = uuid4()
+    Path(f'{dir_name}/img').mkdir(parents=True)
+    im.save(f'{dir_name}/img/image.jpg')
+    os.system(f'python main.py infer {dir_name}/img/image.jpg {dir_name}_output -s f -i')
+    with open(f'{dir_name}_output/image/txt/image_main.txt') as f:
+        return f'{dir_name}_output/image/pred_img/image_L.jpg', f.read()
+
+title = "NDLOCR"
+description = "Gradio demo for NDLOCR. NDLOCR is a text recognition (OCR) Program."
+article = "<p style='text-align: center'><a href='https://github.com/ndl-lab' target='_blank'>NDL Lab</a> | <a href='https://github.com/ndl-lab/ndlocr_cli' target='_blank'>NDLOCR Repo</a></p>"
+gr.Interface(
+    inference, 
+    gr.inputs.Image(label='image', type='pil'), 
+    ['image', 'text'],
+    title=title,
+    description=description,
+    article=article,
+    examples=['fSL1CGG.jpg']
+).launch(enable_queue=True, cache_examples=True)

cli/core/__init__.py

@@ -0,0 +1,9 @@
+# Copyright (c) 2022, National Diet Library, Japan
+#
+# This software is released under the CC BY 4.0.
+# https://creativecommons.org/licenses/by/4.0/
+
+
+from .inference import OcrInferencer
+
+__all__ = ['OcrInferencer']

cli/core/inference.py

@@ -0,0 +1,530 @@
+# Copyright (c) 2022, National Diet Library, Japan
+#
+# This software is released under the CC BY 4.0.
+# https://creativecommons.org/licenses/by/4.0/
+
+
+import copy
+import cv2
+import glob
+import os
+import pathlib
+import sys
+import time
+import xml
+import xml.etree.ElementTree as ET
+
+from . import utils
+from .. import procs
+
+# Add import path for src modules
+currentdir = pathlib.Path(__file__).resolve().parent
+sys.path.append(str(currentdir) + "/../../src/separate_pages_ssd")
+sys.path.append(str(currentdir) + "/../../src/ndl_layout")
+sys.path.append(str(currentdir) + "/../../src/deskew_HT")
+sys.path.append(str(currentdir) + "/../../src/text_recognition")
+
+# supported image type list
+supported_img_ext = ['.jpg', '.jpeg', '.jp2']
+
+
+class OcrInferencer:
+    """
+    推論実行時の関数や推論の設定値を保持します。
+
+    Attributes
+    ----------
+    full_proc_list : list
+        全推論処理のリストです。
+    proc_list : list
+        本実行処理における推論処理のリストです。
+    cfg : dict
+        本実行処理における設定情報です。
+    """
+
+    def __init__(self, cfg):
+        """
+        Parameters
+        ----------
+        cfg : dict
+            本実行処理における設定情報です。
+        """
+        # inference process class list in order
+        self.full_proc_list = [
+            procs.PageSeparation,           # 0: ノド元分割               出力：（画像：あり、XML：なし、TXT：なし）
+            procs.PageDeskewProcess,        # 1: 傾き補正                 出力：（画像：あり、XML：なし、TXT：なし）
+            procs.LayoutExtractionProcess,  # 2: レイアウト抽出           出力：（画像：あり、XML：あり、TXT：なし）
+            procs.LineOcrProcess,           # 3: 文字認識(OCR)            出力：（画像：あり、XML：あり、TXT：あり）
+        ]
+        self.proc_list = self._create_proc_list(cfg)
+        self.cfg = cfg
+        self.time_statistics = []
+        self.xml_template = '<?xml version="1.0" encoding="utf-8" standalone="yes"?>\n<OCRDATASET></OCRDATASET>'
+
+    def run(self):
+        """
+        self.cfgに保存された設定に基づいた推論処理を実行します。
+        """
+        if len(self.cfg['input_dirs']) == 0:
+            print('[ERROR] Input directory list is empty', file=sys.stderr)
+            return
+
+        # input dir loop
+        for input_dir in self.cfg['input_dirs']:
+            if self.cfg['input_structure'] in ['t']:
+                single_outputdir_data_list = self._get_single_dir_data_from_tosho_data(input_dir)
+            else:
+                single_outputdir_data_list = self._get_single_dir_data(input_dir)
+
+            if single_outputdir_data_list is None:
+                print('[ERROR] Input data list is empty', file=sys.stderr)
+                continue
+            print(single_outputdir_data_list)
+            # do infer with input data for single output data dir
+            for single_outputdir_data in single_outputdir_data_list:
+                print(single_outputdir_data)
+                if single_outputdir_data is None:
+                    continue
+                pred_list = self._infer(single_outputdir_data)
+
+                # save inferenced xml in xml directory
+                if (self.cfg['save_xml'] or self.cfg['partial_infer']) and (self.cfg['proc_range']['end'] > 1):
+                    self._save_pred_xml(single_outputdir_data['output_dir'], [single_data['xml'] for single_data in pred_list])
+        if len(self.time_statistics) == 0:
+            print('================== NO VALID INFERENCE ==================')
+        else:
+            average = sum(self.time_statistics) / len(self.time_statistics)
+            print('================== PROCESSING TIME ==================')
+            print('Average processing time : {0} sec / image file '.format(average))
+        return
+
+    def _infer(self, single_outputdir_data):
+        """
+        self.cfgに保存された設定に基づき、XML一つ分のデータに対する推論処理を実行します。
+
+        Parameters
+        ----------
+        single_outputdir_data : dict
+            XML一つ分のデータ（基本的に1書籍分を想定）の入力データ情報。
+            画像ファイルパスのリスト、それらに対応するXMLデータを含みます。
+
+        Returns
+        -------
+        pred_list : list
+            1ページ分の推論結果を要素に持つ推論結果のリスト。
+            各結果は辞書型で保持されています。
+        """
+        # single_outputdir_data dictionary include [key, value] pairs as below
+        # (xml is not always included)
+        #   [key, value]: ['img', numpy.ndarray], ['xml', xml_tree]
+        pred_list = []
+        pred_xml_dict_for_dump = {}
+        if self.cfg['dump']:
+            dump_dir = os.path.join(single_outputdir_data['output_dir'], 'dump')
+            os.makedirs(dump_dir, exist_ok=True)
+
+            for proc in self.proc_list:
+                pred_xml_dict_for_dump[proc.proc_name] = []
+                proc_dump_dir = os.path.join(dump_dir, proc.proc_name)
+                os.makedirs(proc_dump_dir, exist_ok=True)
+
+        for img_path in single_outputdir_data['img_list']:
+            single_image_file_data = self._get_single_image_file_data(img_path, single_outputdir_data)
+            output_dir = single_outputdir_data['output_dir']
+            if single_image_file_data is None:
+                print('[ERROR] Failed to get single page input data for image:{0}'.format(img_path), file=sys.stderr)
+                continue
+
+            print('######## START PAGE INFERENCE PROCESS ########')
+            start_page = time.time()
+
+            for proc in self.proc_list:
+                single_page_output = []
+                for idx, single_data_input in enumerate(single_image_file_data):
+                    single_data_output = proc.do(idx, single_data_input)
+                    single_page_output.extend(single_data_output)
+                # save inference result data to dump
+                if self.cfg['dump'] and 'xml' in single_image_file_data[0].keys():
+                    pred_xml_dict_for_dump[proc.proc_name].append(single_image_file_data[0]['xml'])
+
+                single_image_file_data = single_page_output
+
+            single_image_file_output = single_image_file_data
+            self.time_statistics.append(time.time() - start_page)
+
+            if self.cfg['save_image'] or self.cfg['partial_infer']:
+                # save inferenced result drawn image in pred_img directory
+                for single_data_output in single_image_file_output:
+                    # save input image while partial inference
+                    if self.cfg['partial_infer']:
+                        img_output_dir = os.path.join(output_dir, 'img')
+                        self._save_image(single_data_output['img'], single_data_output['img_file_name'], img_output_dir)
+
+                    pred_img = self._create_result_image(single_data_output, self.proc_list[-1].proc_name)
+                    img_output_dir = os.path.join(output_dir, 'pred_img')
+                    self._save_image(pred_img, single_data_output['img_file_name'], img_output_dir)
+
+            # save inferenced result text for this page
+            if self.cfg['proc_range']['end'] > 2:
+                sum_main_txt = ''
+                sum_cap_txt = ''
+                for single_data_output in single_image_file_output:
+                    main_txt, cap_txt = self._create_result_txt(single_data_output['xml'])
+                    sum_main_txt += main_txt + '\n'
+                    sum_cap_txt += sum_cap_txt + '\n'
+                self._save_pred_txt(sum_main_txt, sum_cap_txt, os.path.basename(img_path), single_outputdir_data['output_dir'])
+
+            # add inference result for single image file data to pred_list, including XML data
+            pred_list.extend(single_image_file_output)
+            print('########  END PAGE INFERENCE PROCESS  ########')
+
+        return pred_list
+
+    def _get_single_dir_data(self, input_dir):
+        """
+        XML一つ分の入力データに関する情報を整理して取得します。
+
+        Parameters
+        ----------
+        input_dir : str
+            XML一つ分の入力データが保存されているディレクトリパスです。
+
+        Returns
+        -------
+        # Fixme
+        single_dir_data : dict
+            XML一つ分のデータ（基本的に1PID分を想定）の入力データ情報です。
+            画像ファイルパスのリスト、それらに対応するXMLデータを含みます。
+        """
+        single_dir_data = {'input_dir': os.path.abspath(input_dir)}
+        single_dir_data['img_list'] = []
+
+        # get img list of input directory
+        if self.cfg['input_structure'] in ['w']:
+            for ext in supported_img_ext:
+                single_dir_data['img_list'].extend(sorted(glob.glob(os.path.join(input_dir, '*{0}'.format(ext)))))
+        elif self.cfg['input_structure'] in ['f']:
+            stem, ext = os.path.splitext(os.path.basename(input_dir))
+            if ext in supported_img_ext:
+                single_dir_data['img_list'] = [input_dir]
+            else:
+                print('[ERROR] This file is not supported type : {0}'.format(input_dir), file=sys.stderr)
+        elif not os.path.isdir(os.path.join(input_dir, 'img')):
+            print('[ERROR] Input img diretctory not found in {}'.format(input_dir), file=sys.stderr)
+            return None
+        else:
+            for ext in supported_img_ext:
+                single_dir_data['img_list'].extend(sorted(glob.glob(os.path.join(input_dir, 'img/*{0}'.format(ext)))))
+
+        # check xml file number and load xml data if needed
+        if self.cfg['proc_range']['start'] > 2:
+            if self.cfg['input_structure'] in ['f']:
+                print('[ERROR] Single image file input mode does not support partial inference wich need xml file input.', file=sys.stderr)
+                return None
+            input_xml = None
+            xml_file_list = glob.glob(os.path.join(input_dir, 'xml/*.xml'))
+            if len(xml_file_list) > 1:
+                print('[ERROR] Input xml file must be only one, but there is {0} xml files in {1}.'.format(
+                    len(xml_file_list), os.path.join(self.cfg['input_root'], 'xml')), file=sys.stderr)
+                return None
+            elif len(xml_file_list) == 0:
+                print('[ERROR] There is no input xml files in {0}.'.format(os.path.join(input_dir, 'xml')), file=sys.stderr)
+                return None
+            else:
+                input_xml = xml_file_list[0]
+            try:
+                single_dir_data['xml'] = ET.parse(input_xml)
+            except xml.etree.ElementTree.ParseError as err:
+                print("[ERROR] XML parse error : {0}".format(input_xml), file=sys.stderr)
+                return None
+
+        # prepare output dir for inferensce result with this input dir
+        if self.cfg['input_structure'] in ['f']:
+            stem, ext = os.path.splitext(os.path.basename(input_dir))
+            output_dir = os.path.join(self.cfg['output_root'], stem)
+        elif self.cfg['input_structure'] in ['i', 's']:
+            dir_name = os.path.basename(input_dir)
+            output_dir = os.path.join(self.cfg['output_root'], dir_name)
+        elif self.cfg['input_structure'] in ['w']:
+            input_dir_names = input_dir.split('/')
+            dir_name = input_dir_names[-3][0] + input_dir_names[-2] + input_dir_names[-1]
+            output_dir = os.path.join(self.cfg['output_root'], dir_name)
+        else:
+            print('[ERROR] Unexpected input directory structure type: {}.'.format(self.cfg['input_structure']), file=sys.stderr)
+            return None
+
+        # output directory existance check
+        output_dir = utils.mkdir_with_duplication_check(output_dir)
+        single_dir_data['output_dir'] = output_dir
+
+        return [single_dir_data]
+
+    def _get_single_dir_data_from_tosho_data(self, input_dir):
+        """
+        XML一つ分の入力データに関する情報を整理して取得します。
+
+        Parameters
+        ----------
+        input_dir : str
+            tosho data形式のセクションごとのディレクトリパスです。
+
+        Returns
+        -------
+        single_dir_data_list : list
+            XML一つ分のデータ（基本的に1PID分を想定）の入力データ情報のリストです。
+            1つの要素に画像ファイルパスのリスト、それらに対応するXMLデータを含みます。
+        """
+        single_dir_data_list = []
+
+        # get img list of input directory
+        tmp_img_list = sorted(glob.glob(os.path.join(input_dir, '*.jp2')))
+        tmp_img_list.extend(sorted(glob.glob(os.path.join(input_dir, '*.jpg'))))
+
+        pid_list = []
+        for img in tmp_img_list:
+            pid = os.path.basename(img).split('_')[0]
+            if pid not in pid_list:
+                pid_list.append(pid)
+
+        for pid in pid_list:
+            single_dir_data = {'input_dir': os.path.abspath(input_dir),
+                               'img_list': [img for img in tmp_img_list if os.path.basename(img).startswith(pid)]}
+
+            # prepare output dir for inferensce result with this input dir
+            output_dir = os.path.join(self.cfg['output_root'], pid)
+
+            # output directory existance check
+            os.makedirs(output_dir, exist_ok=True)
+            single_dir_data['output_dir'] = output_dir
+            single_dir_data_list.append(single_dir_data)
+
+        return single_dir_data_list
+
+    def _get_single_image_file_data(self, img_path, single_dir_data):
+        """
+        1ページ分の入力データに関する情報を整理して取得します。
+
+        Parameters
+        ----------
+        img_path : str
+            入力画像データのパスです。
+        single_dir_data : dict
+            1書籍分の入力データに関する情報を保持する辞書型データです。
+            xmlファイルへのパス、結果を出力するディレクトリのパスなどを含みます。
+
+        Returns
+        -------
+        single_image_file_data : dict
+            1ページ分のデータの入力データ情報です。
+            画像ファイルのパスとnumpy.ndarray形式の画像データ、その画像に対応するXMLデータを含みます。
+        """
+        single_image_file_data = [{
+            'img_path': img_path,
+            'img_file_name': os.path.basename(img_path),
+            'output_dir': single_dir_data['output_dir']
+        }]
+
+        full_xml = None
+        if 'xml' in single_dir_data.keys():
+            full_xml = single_dir_data['xml']
+
+        # get img data for single page
+        orig_img = cv2.imread(img_path)
+        if orig_img is None:
+            print('[ERROR] Image read error : {0}'.format(img_path), file=sys.stderr)
+            return None
+        single_image_file_data[0]['img'] = orig_img
+
+        # return if this proc needs only img data for input
+        if full_xml is None:
+            return single_image_file_data
+
+        # get xml data for single page
+        image_name = os.path.basename(img_path)
+        for page in full_xml.getroot().iter('PAGE'):
+            if page.attrib['IMAGENAME'] == image_name:
+                node = ET.fromstring(self.xml_template)
+                node.append(page)
+                tree = ET.ElementTree(node)
+                single_image_file_data[0]['xml'] = tree
+                break
+
+        # [TODO] 画像データに対応するXMLデータが見つからなかった場合の対応
+        if 'xml' not in single_image_file_data[0].keys():
+            print('[ERROR] Input XML data for page {} not found.'.format(img_path), file=sys.stderr)
+
+        return single_image_file_data
+
+    def _create_proc_list(self, cfg):
+        """
+        推論の設定情報に基づき、実行する推論処理のリストを作成します。
+
+        Parameters
+        ----------
+        cfg : dict
+            推論実行時の設定情報を保存した辞書型データ。
+        """
+        proc_list = []
+        for i in range(cfg['proc_range']['start'], cfg['proc_range']['end'] + 1):
+            proc_list.append(self.full_proc_list[i](cfg, i))
+        return proc_list
+
+    def _save_pred_xml(self, output_dir, pred_list):
+        """
+        推論結果のXMLデータをまとめたXMLファイルを生成して保存します。
+
+        Parameters
+        ----------
+        output_dir : str
+            推論結果を保存するディレクトリのパスです。
+        pred_list : list
+            1ページ分の推論結果を要素に持つ推論結果のリスト。
+            各結果は辞書型で保持されています。
+        """
+        xml_dir = os.path.join(output_dir, 'xml')
+        os.makedirs(xml_dir, exist_ok=True)
+
+        # basically, output_dir is supposed to be PID, so it used as xml filename
+        xml_path = os.path.join(xml_dir, '{}.xml'.format(os.path.basename(output_dir)))
+        pred_xml = self._parse_pred_list_to_save(pred_list)
+        utils.save_xml(pred_xml, xml_path)
+        return
+
+    def _save_image(self, pred_img, orig_img_name, img_output_dir, id=''):
+        """
+        指定されたディレクトリに画像データを保存します。
+        画像データは入力に使用したものと推論結果を重畳したものの２種類が想定されています。
+
+        Parameters
+        ----------
+        pred_img : numpy.ndarray
+            保存する画像データ。
+        orig_img_name : str
+            もともとの入力画像のファイル名。
+            基本的にはこのファイル名と同名で保存します。
+        img_output_dir : str
+            画像ファイルの保存先のディレクトリパス。
+        id : str
+            もともとの入力画像のファイル名に追加する処理結果ごとのidです。
+            一つの入力画像から複数の画像データが出力される処理がある場合に必要になります。
+        """
+        os.makedirs(img_output_dir, exist_ok=True)
+        stem, ext = os.path.splitext(orig_img_name)
+        orig_img_name = stem + '.jpg'
+
+        if id != '':
+            stem, ext = os.path.splitext(orig_img_name)
+            orig_img_name = stem + '_' + id + ext
+
+        img_path = os.path.join(img_output_dir, orig_img_name)
+        try:
+            cv2.imwrite(img_path, pred_img)
+        except OSError as err:
+            print("[ERROR] Image save error: {0}".format(err), file=sys.stderr)
+            raise OSError
+
+        return
+
+    def _save_pred_txt(self, main_txt, cap_txt, orig_img_name, output_dir):
+        """
+        指定されたディレクトリに推論結果のテキストデータを保存します。
+
+        Parameters
+        ----------
+        main_txt : str
+            本文＋キャプションの推論結果のテキストデータです
+        cap_txt : str
+            キャプションのみの推論結果のテキストデータです
+        orig_img_name : str
+            もともとの入力画像ファイル名。
+            基本的にはこのファイル名と同名で保存します。
+        img_output_dir : str
+            画像ファイルの保存先のディレクトリパス。
+        """
+        txt_dir = os.path.join(output_dir, 'txt')
+        os.makedirs(txt_dir, exist_ok=True)
+
+        stem, _ = os.path.splitext(orig_img_name)
+        txt_path = os.path.join(txt_dir, stem + '_cap.txt')
+        try:
+            with open(txt_path, 'w') as f:
+                f.write(cap_txt)
+        except OSError as err:
+            print("[ERROR] Caption text save error: {0}".format(err), file=sys.stderr)
+            raise OSError
+
+        stem, _ = os.path.splitext(orig_img_name)
+        txt_path = os.path.join(txt_dir, stem + '_main.txt')
+        try:
+            with open(txt_path, 'w') as f:
+                f.write(main_txt)
+        except OSError as err:
+            print("[ERROR] Main text save error: {0}".format(err), file=sys.stderr)
+            raise OSError
+
+        return
+
+    def _parse_pred_list_to_save(self, pred_list):
+        """
+        推論結果のXMLを要素に持つリストから、ファイルに保存するための一つのXMLデータを生成します。
+
+        Parameters
+        ----------
+        pred_list : list
+            推論結果のXMLを要素に持つリスト。
+        """
+        ET.register_namespace('', 'NDLOCRDATASET')
+        node = ET.fromstring(self.xml_template)
+        for single_xml_tree in pred_list:
+            root = single_xml_tree.getroot()
+            for element in root:
+                node.append(element)
+
+        tree = ET.ElementTree(node)
+        return tree
+
+    def _create_result_image(self, result, proc_name):
+        """
+        推論結果を入力画像に重畳した画像データを生成します。
+
+        Parameters
+        ----------
+        result : dict
+            1ページ分の推論結果を持つ辞書型データ。
+        proc_name : str
+            重畳を行う結果を出力した推論処理の名前。
+        """
+        if 'dump_img' in result.keys():
+            dump_img = copy.deepcopy(result['dump_img'])
+        else:
+            dump_img = copy.deepcopy(result['img'])
+        if 'xml' in result.keys() and result['xml'] is not None:
+            # draw inference result on input image
+            cv2.putText(dump_img, proc_name, (0, 50),
+                        cv2.FONT_HERSHEY_PLAIN, 4, (0, 0, 0), 5, cv2.LINE_AA)
+            pass
+        else:
+            cv2.putText(dump_img, proc_name, (0, 50),
+                        cv2.FONT_HERSHEY_PLAIN, 4, (0, 0, 0), 5, cv2.LINE_AA)
+        return dump_img
+
+    def _create_result_txt(self, xml_data):
+        """
+        推論結果のxmlデータからテキストデータを生成します。
+
+        Parameters
+        ----------
+        xml_data :
+            1ページ分の推論結果を持つxmlデータ。
+        """
+        main_txt = ''
+        cap_txt = ''
+        for page_xml in xml_data.iter('PAGE'):
+            for line_xml in page_xml.iter('LINE'):
+                main_txt += line_xml.attrib['STRING']
+                main_txt += '\n'
+                if line_xml.attrib['TYPE'] == 'キャプション':
+                    cap_txt += line_xml.attrib['STRING']
+                    cap_txt += '\n'
+
+        return main_txt, cap_txt

cli/core/utils.py

@@ -0,0 +1,195 @@
+# Copyright (c) 2022, National Diet Library, Japan
+#
+# This software is released under the CC BY 4.0.
+# https://creativecommons.org/licenses/by/4.0/
+
+
+import copy
+import datetime
+import glob
+import os
+import sys
+import yaml
+
+
+def parse_cfg(cfg_dict):
+    """
+    コマンドで入力された引数やオプションを内部関数が利用しやすい形にparseします。
+
+    Parameters
+    ----------
+    cfg_dict : dict
+        コマンドで入力された引数やオプションが保存された辞書型データ。
+
+    Returns
+    -------
+    infer_cfg : dict
+        推論処理を実行するための設定情報が保存された辞書型データ。
+    """
+    infer_cfg = copy.deepcopy(cfg_dict)
+
+    # add inference config parameters from yml config file
+    yml_config = None
+    if not os.path.isfile(cfg_dict['config_file']):
+        print('[ERROR] Config yml file not found.', file=sys.stderr)
+        return None
+
+    with open(cfg_dict['config_file'], 'r') as yml:
+        yml_config = yaml.safe_load(yml)
+
+    if type(yml_config) is not dict:
+        print('[ERROR] Config yml file read error.', file=sys.stderr)
+        return None
+
+    infer_cfg.update(yml_config)
+
+    # save_xml will be ignored when last proc does not output xml data
+    if (infer_cfg['proc_range'] != '0..3') and (infer_cfg['save_xml'] or infer_cfg['save_image']):
+        print('[WARNING] save_xml and save_image flags are ignored because this is partial execution.')
+        print('          All output of last proc will be saved in output directory.')
+
+    # parse start/end indices of inference process
+    start = int(infer_cfg['proc_range'][0])
+    end = int(infer_cfg['proc_range'][-1])
+    if start > end:
+        print('[ERROR] Value of proc_range must be [x..y : x <= y] .', file=sys.stderr)
+        return None
+    infer_cfg['proc_range'] = {
+        'start': start,
+        'end': end
+    }
+    if (start != 0) or (end != 3):
+        infer_cfg['partial_infer'] = True
+    else:
+        infer_cfg['partial_infer'] = False
+
+    # create input_dirs from input_root
+    # input_dirs is list of dirs that contain img (and xml) dir
+    infer_cfg['input_root'] = os.path.abspath(infer_cfg['input_root'])
+    infer_cfg['output_root'] = os.path.abspath(infer_cfg['output_root'])
+    if infer_cfg['input_structure'] in ['s']:
+        # - Sigle input dir mode
+        # input_root
+        #  ├── xml
+        #  │   └── R[7桁連番].xml※XMLデータ
+        #  └── img
+        #      └── R[7桁連番]_pp.jp2※画像データ
+
+        # validation check for input dir structure
+        if not os.path.isdir(os.path.join(infer_cfg['input_root'], 'img')):
+            print('[ERROR] Input img diretctory not found in {}'.format(infer_cfg['input_root']), file=sys.stderr)
+            return None
+        if (start > 2) and (not os.path.isdir(os.path.join(infer_cfg['input_root'], 'xml'))):
+            print('[ERROR] Input xml diretctory not found in {}'.format(infer_cfg['input_root']), file=sys.stderr)
+            return None
+        infer_cfg['input_dirs'] = [infer_cfg['input_root']]
+    elif infer_cfg['input_structure'] in ['i']:
+        # - Partial inference mode
+        # input_root
+        #  └── PID
+        #      ├── xml
+        #      │   └── R[7桁連番].xml※XMLデータ
+        #      └── img
+        #          └── R[7桁連番]_pp.jp2※画像データ
+        infer_cfg['input_dirs'] = []
+        for input_dir in glob.glob(os.path.join(infer_cfg['input_root'], '*')):
+            if os.path.isdir(input_dir):
+                if not os.path.isdir(os.path.join(input_dir, 'img')):
+                    print('[WARNING] Input directory {0} is skipped(no img diretctory)'.format(input_dir))
+                    continue
+                if (start > 1) and (not os.path.isdir(os.path.join(input_dir, 'xml'))):
+                    print('[WARNING] Input directory {0} is skipped(no xml diretctory)'.format(input_dir))
+                    continue
+                infer_cfg['input_dirs'].append(input_dir)
+    elif infer_cfg['input_structure'] in ['t']:
+        # - ToshoData mode
+        # input_root
+        #  └── tosho_19XX_bunkei
+        #      └── R[7桁連番]_pp.jp2※画像データ
+        infer_cfg['input_dirs'] = []
+        for input_dir in glob.glob(os.path.join(infer_cfg['input_root'], '*')):
+            if os.path.isdir(input_dir):
+                infer_cfg['input_dirs'].append(input_dir)
+        if 'img' in [os.path.basename(d) for d in infer_cfg['input_dirs']]:
+            print('[WARNING] This input structure might be single input(img diretctory found)')
+    elif infer_cfg['input_structure'] in ['w']:
+        # - Work station input mode
+        # input_root
+        #  └── workstation
+        #      └── [collect(3桁数字)、またはdigital(3桁数字)]フォルダ
+        #           └── [15桁連番]フォルダ※PID上1桁目
+        #                └── [3桁連番]フォルダ※PID上2～4桁目
+        #                     └── [3桁連番]フォルダ※PID上5～7桁目
+        #                          └── R[7桁連番]_contents.jp2※画像データ
+
+        # recursive function to get input_dirs in workstation mode
+        def get_input_dirs(path, depth):
+            depth += 1
+            ret_list = []
+            current_list = []
+            for input_dir in glob.glob(os.path.join(path, '*')):
+                if os.path.isdir(input_dir):
+                    current_list.append(input_dir)
+            if depth > 3:
+                return current_list
+            if (depth < 2) and (len(current_list) == 0):
+                print('[ERROR] Input directory structure dose not match workstation mode', file=sys.stderr)
+                return []
+            for dir in current_list:
+                tmp_list = get_input_dirs(dir, depth)
+                ret_list.extend(tmp_list)
+            return ret_list
+
+        # check if workstation directory exist
+        work_dir = os.path.join(infer_cfg['input_root'], 'workstation')
+        if not os.path.isdir(work_dir):
+            print('[ERROR] \'workstation\' directory not found', file=sys.stderr)
+            return None
+
+        # get input dir list
+        infer_cfg['input_dirs'] = get_input_dirs(work_dir, 0)
+    elif infer_cfg['input_structure'] in ['f']:
+        # - Image file input mode
+        # input_root is equal to input image file path
+        infer_cfg['input_dirs'] = [infer_cfg['input_root']]
+    else:
+        print('[ERROR] Unexpected input directory structure type: {0}.'.format(infer_cfg['input_structure']), file=sys.stderr)
+        return None
+
+    return infer_cfg
+
+
+def save_xml(xml_to_save, path):
+    """
+    指定されたファイルパスにXMLファイル保存します。
+
+    Parameters
+    ----------
+    path : str
+        XMLファイルを保存するファイルパス。
+
+    """
+    print('### save xml : {}###'.format(path))
+    try:
+        xml_to_save.write(path, encoding='utf-8', xml_declaration=True)
+    except OSError as err:
+        print("[ERROR] XML save error : {0}".format(err), file=sys.stderr)
+        raise OSError
+    return
+
+
+def mkdir_with_duplication_check(dir_path):
+    dir_path_to_create = dir_path
+
+    # prepare output root derectory
+    while os.path.isdir(dir_path_to_create):
+        print('[WARNING] Directory {0} already exist.'.format(dir_path))
+        now = datetime.datetime.now()
+        time_stamp = now.strftime('_%Y%m%d%H%M%S')
+        dir_path_to_create += time_stamp
+
+    if dir_path_to_create != dir_path:
+        print('[WARNING] Directory is changed to {0}.'.format(dir_path_to_create))
+    os.mkdir(dir_path_to_create)
+
+    return dir_path_to_create

cli/procs/__init__.py

@@ -0,0 +1,12 @@
+# Copyright (c) 2022, National Diet Library, Japan
+#
+# This software is released under the CC BY 4.0.
+# https://creativecommons.org/licenses/by/4.0/
+
+
+from .page_separation import PageSeparation
+from .page_deskew import PageDeskewProcess
+from .layout_extraction import LayoutExtractionProcess
+from .line_ocr import LineOcrProcess
+
+__all__ = ['PageSeparation', 'PageDeskewProcess', 'LayoutExtractionProcess', 'LineOcrProcess']

cli/procs/base_proc.py

@@ -0,0 +1,275 @@
+# Copyright (c) 2022, National Diet Library, Japan
+#
+# This software is released under the CC BY 4.0.
+# https://creativecommons.org/licenses/by/4.0/
+
+
+import copy
+import cv2
+import os
+
+
+class BaseInferenceProcess:
+    """
+    各推論処理を実行するプロセスクラスを作るためのメタクラス。
+
+    Attributes
+    ----------
+    proc_name : str
+        推論処理を実行するインスタンスが持つプロセス名。
+        [実行される順序を表す数字＋クラスごとの処理名]で構成されます。
+    cfg : dict
+        本推論実行における設定情報です。
+    """
+    def __init__(self, cfg, pid, proc_type='_base_prep'):
+        """
+        Parameters
+        ----------
+        cfg : dict
+            本実行処理における設定情報です。
+        pid : int
+            実行される順序を表す数値。
+        proc_type : str
+            クラスごとに定義されている処理名。
+        """
+        self.proc_name = str(pid) + proc_type
+
+        if not self._is_valid_cfg(cfg):
+            raise ValueError('Configuration validation error.')
+        else:
+            self.cfg = cfg
+
+        self.process_dump_dir = None
+
+        return True
+
+    def do(self, data_idx, input_data):
+        """
+        推論処理を実行する際にOcrInferencerクラスから呼び出される推論実行関数。
+        入力データのバリデーションや推論処理、推論結果の保存などが含まれます。
+        本処理は基本的に継承先では変更されないことを想定しています。
+
+        Parameters
+        ----------
+        data_idx : int
+            入力データのインデックス。
+            画像ファイル１つごとに入力データのリストが構成されます。
+        input_data : dict
+            推論処理を実行すつ対象の入力データ。
+
+        Returns
+        -------
+        result : dict
+            推論処理の結果を保持する辞書型データ。
+            基本的にinput_dataと同じ構造です。
+        """
+        # input data valudation check
+        if not self._is_valid_input(input_data):
+            raise ValueError('Input data validation error.')
+
+        # run main inference process
+        result = self._run_process(input_data)
+        if result is None:
+            raise ValueError('Inference output error in {0}.'.format(self.proc_name))
+
+        # dump inference result
+        if self.cfg['dump']:
+            self._dump_result(input_data, result, data_idx)
+
+        return result
+
+    def _run_process(self, input_data):
+        """
+        推論処理の本体部分。
+        処理内容は継承先のクラスで実装されることを想定しています。
+
+        Parameters
+        ----------
+        input_data : dict
+            推論処理を実行する対象の入力データ。
+
+        Returns
+        -------
+        result : dict
+            推論処理の結果を保持する辞書型データ。
+            基本的にinput_dataと同じ構造です。
+        """
+        print('### Base Inference Process ###')
+        result = copy.deepcopy(input_data)
+        return result
+
+    def _is_valid_cfg(self, cfg):
+        """
+        推論処理全体の設定情報ではなく、クラス単位の設定情報に対するバリデーション。
+        バリデーションの内容は継承先のクラスで実装されることを想定しています。
+
+        Parameters
+        ----------
+        cfg : dict
+            本推論実行における設定情報です。
+
+        Returns
+        -------
+        [変数なし] : bool
+            設定情報が正しければTrue, そうでなければFalseを返します。
+        """
+        if cfg is None:
+            print('Given configuration data is None.')
+            return False
+        return True
+
+    def _is_valid_input(self, input_data):
+        """
+        本クラスの推論処理における入力データのバリデーション。
+        バリデーションの内容は継承先のクラスで実装されることを想定しています。
+
+        Parameters
+        ----------
+        input_data : dict
+            推論処理を実行する対象の入力データ。
+
+        Returns
+        -------
+        [変数なし] : bool
+            　入力データが正しければTrue, そうでなければFalseを返します。
+        """
+        return True
+
+    def _dump_result(self, input_data, result, data_idx):
+        """
+        本クラスの推論処理結果をファイルに保存します。
+        dumpフラグが有効の場合にのみ実行されます。
+
+        Parameters
+        ----------
+        input_data : dict
+            推論処理に利用した入力データ。
+        result : list
+            推論処理の結果を保持するリスト型データ。
+            各要素は基本的にinput_dataと同じ構造の辞書型データです。
+        data_idx : int
+            入力データのインデックス。
+            画像ファイル１つごとに入力データのリストが構成されます。
+        """
+
+        self.process_dump_dir = os.path.join(os.path.join(input_data['output_dir'], 'dump'), self.proc_name)
+
+        for i, single_result in enumerate(result):
+            if 'img' in single_result.keys() and single_result['img'] is not None:
+                dump_img_name = os.path.basename(input_data['img_path']).split('.')[0] + '_' + str(data_idx) + '_' + str(i) + '.jpg'
+                self._dump_img_result(single_result, input_data['output_dir'], dump_img_name)
+            if 'xml' in single_result.keys() and single_result['xml'] is not None:
+                dump_xml_name = os.path.basename(input_data['img_path']).split('.')[0] + '_' + str(data_idx) + '_' + str(i) + '.xml'
+                self._dump_xml_result(single_result, input_data['output_dir'], dump_xml_name)
+            if 'txt' in single_result.keys() and single_result['txt'] is not None:
+                dump_txt_name = os.path.basename(input_data['img_path']).split('.')[0] + '_' + str(data_idx) + '_' + str(i) + '.txt'
+                self._dump_txt_result(single_result, input_data['output_dir'], dump_txt_name)
+        return
+
+    def _dump_img_result(self, single_result, output_dir, img_name):
+        """
+        本クラスの推論処理結果(画像)をファイルに保存します。
+        dumpフラグが有効の場合にのみ実行されます。
+
+        Parameters
+        ----------
+        single_result : dict
+            推論処理の結果を保持する辞書型データ。
+        output_dir : str
+            推論結果が保存されるディレクトリのパス。
+        img_name : str
+            入力データの画像ファイル名。
+            dumpされる画像ファイルのファイル名は入力のファイル名と同名(複数ある場合は連番を付与)となります。
+        """
+        pred_img_dir = os.path.join(self.process_dump_dir, 'pred_img')
+        os.makedirs(pred_img_dir, exist_ok=True)
+        image_file_path = os.path.join(pred_img_dir, img_name)
+        dump_image = self._create_result_image(single_result)
+        try:
+            cv2.imwrite(image_file_path, dump_image)
+        except OSError as err:
+            print("Dump image save error: {0}".format(err))
+            raise OSError
+
+        return
+
+    def _dump_xml_result(self, single_result, output_dir, img_name):
+        """
+        本クラスの推論処理結果(XML)をファイルに保存します。
+        dumpフラグが有効の場合にのみ実行されます。
+
+        Parameters
+        ----------
+        single_result : dict
+            推論処理の結果を保持する辞書型データ。
+        output_dir : str
+            推論結果が保存されるディレクトリのパス。
+        img_name : str
+            入力データの画像ファイル名。
+            dumpされるXMLファイルのファイル名は入力のファイル名とほぼ同名（拡張子の変更、サフィックスや連番の追加のみ）となります。
+        """
+        xml_dir = os.path.join(self.process_dump_dir, 'xml')
+        os.makedirs(xml_dir, exist_ok=True)
+        trum, _ = os.path.splitext(img_name)
+        xml_path = os.path.join(xml_dir, trum + '.xml')
+        try:
+            single_result['xml'].write(xml_path, encoding='utf-8', xml_declaration=True)
+        except OSError as err:
+            print("Dump xml save error: {0}".format(err))
+            raise OSError
+
+        return
+
+    def _dump_txt_result(self, single_result, output_dir, img_name):
+        """
+        本クラスの推論処理結果(テキスト)をファイルに保存します。
+        dumpフラグが有効の場合にのみ実行されます。
+
+        Parameters
+        ----------
+        single_result : dict
+            推論処理の結果を保持する辞書型データ。
+        output_dir : str
+            推論結果が保存されるディレクトリのパス。
+        img_name : str
+            入力データの画像ファイル名。
+            dumpされるテキストファイルのファイル名は入力のファイル名とほぼ同名（拡張子の変更、サフィックスや連番の追加のみ）となります。
+        """
+        txt_dir = os.path.join(self.process_dump_dir, 'txt')
+        os.makedirs(txt_dir, exist_ok=True)
+
+        trum, _ = os.path.splitext(img_name)
+        txt_path = os.path.join(txt_dir, trum + '_main.txt')
+        try:
+            with open(txt_path, 'w') as f:
+                f.write(single_result['txt'])
+        except OSError as err:
+            print("Dump text save error: {0}".format(err))
+            raise OSError
+
+        return
+
+    def _create_result_image(self, single_result):
+        """
+        ���論結果を入力の画像に重畳した画像データを生成します。
+
+        Parameters
+        ----------
+        single_result : dict
+            推論処理の結果を保持する辞書型データ。
+        """
+        dump_img = None
+        if 'dump_img' in single_result.keys():
+            dump_img = copy.deepcopy(single_result['dump_img'])
+        else:
+            dump_img = copy.deepcopy(single_result['img'])
+        if 'xml' in single_result.keys() and single_result['xml'] is not None:
+            # draw single inferenceresult on input image
+            # this should be implemeted in each child class
+            cv2.putText(dump_img, 'dump' + self.proc_name, (0, 50),
+                        cv2.FONT_HERSHEY_PLAIN, 4, (255, 0, 0), 5, cv2.LINE_AA)
+            pass
+        else:
+            cv2.putText(dump_img, 'dump' + self.proc_name, (0, 50),
+                        cv2.FONT_HERSHEY_PLAIN, 4, (255, 255, 0), 5, cv2.LINE_AA)
+        return dump_img

cli/procs/layout_extraction.py

@@ -0,0 +1,79 @@
+# Copyright (c) 2022, National Diet Library, Japan
+#
+# This software is released under the CC BY 4.0.
+# https://creativecommons.org/licenses/by/4.0/
+
+
+import copy
+import numpy
+
+from .base_proc import BaseInferenceProcess
+
+
+class LayoutExtractionProcess(BaseInferenceProcess):
+    """
+    レイアウト抽出推論を実行するプロセスのクラス。
+    BaseInferenceProcessを継承しています。
+    """
+    def __init__(self, cfg, pid):
+        """
+        Parameters
+        ----------
+        cfg : dict
+            本実行処理における設定情報です。
+        pid : int
+            実行される順序を表す数値。
+        """
+        super().__init__(cfg, pid, '_layer_ext')
+        from src.ndl_layout.tools.process import InferencerWithCLI
+        self._inferencer = InferencerWithCLI(self.cfg['layout_extraction'])
+        self._run_src_inference = self._inferencer.inference_wich_cli
+
+    def is_valid_input(self, input_data):
+        """
+        本クラスの推論処理における入力データのバリデーション。
+
+        Parameters
+        ----------
+        input_data : dict
+            推論処理を実行する対象の入力データ。
+
+        Returns
+        -------
+        [変数なし] : bool
+            　入力データが正しければTrue, そうでなければFalseを返します。
+        """
+        if type(input_data['img']) is not numpy.ndarray:
+            print('LayoutExtractionProcess: input img is not numpy.ndarray')
+            return False
+        return True
+
+    def _run_process(self, input_data):
+        """
+        推論処理の本体部分。
+
+        Parameters
+        ----------
+        input_data : dict
+            推論処理を実行する対象の入力データ。
+
+        Returns
+        -------
+        result : dict
+            推論処理の結果を保持する辞書型データ。
+            基本的にinput_dataと同じ構造です。
+        """
+        print('### Layout Extraction Process ###')
+        output_data = copy.deepcopy(input_data)
+        inference_output = self._run_src_inference(img=input_data['img'],
+                                                   img_path=input_data['img_file_name'],
+                                                   score_thr=self.cfg['layout_extraction']['score_thr'],
+                                                   dump=(self.cfg['dump'] or self.cfg['save_image']))
+
+        # Create result to pass xml and img data
+        result = []
+        output_data['xml'] = inference_output['xml']
+        if inference_output['dump_img'] is not None:
+            output_data['dump_img'] = inference_output['dump_img']
+        result.append(output_data)
+        return result

cli/procs/line_ocr.py

@@ -0,0 +1,86 @@
+# Copyright (c) 2022, National Diet Library, Japan
+#
+# This software is released under the CC BY 4.0.
+# https://creativecommons.org/licenses/by/4.0/
+
+
+import copy
+import numpy
+import subprocess
+import xml.etree.ElementTree as ET
+
+from .base_proc import BaseInferenceProcess
+
+
+class LineOcrProcess(BaseInferenceProcess):
+    """
+    行文字認識推論を実行するプロセスのクラス。
+    BaseInferenceProcessを継承しています。
+    """
+    def __init__(self, cfg, pid):
+        """
+        Parameters
+        ----------
+        cfg : dict
+            本推論処理における設定情報です。
+        pid : int
+            実行される順序を表す数値。
+        """
+        super().__init__(cfg, pid, '_line_ocr')
+        process1 = subprocess.Popen(['cat', self.cfg['line_ocr']['char_list']], stdout=subprocess.PIPE)
+        process2 = subprocess.Popen(['tr', '-d', '\\n'], stdin=process1.stdout, stdout=subprocess.PIPE)
+        self.character = '〓' + process2.stdout.read().decode()
+
+        from src.text_recognition.text_recognition import InferencerWithCLI
+        self._inferencer = InferencerWithCLI(self.cfg['line_ocr'], self.character)
+        self._run_src_inference = self._inferencer.inference_wich_cli
+
+    def _is_valid_input(self, input_data):
+        """
+        本クラスの推論処理における入力データのバリデーション。
+
+        Parameters
+        ----------
+        input_data : dict
+            推論処理を実行する対象の入力データ。
+
+        Returns
+        -------
+        [変数なし] : bool
+            入力データが正しければTrue, そうでなければFalseを返します。
+        """
+        if type(input_data['img']) is not numpy.ndarray:
+            print('LineOcrProcess: input img is not numpy.ndarray')
+            return False
+        if type(input_data['xml']) is not ET.ElementTree:
+            print('LineOcrProcess: input xml is not ElementTree')
+            return False
+        return True
+
+    def _run_process(self, input_data):
+        """
+        推論処理の本体部分。
+
+        Parameters
+        ----------
+        input_data : dict
+            推論処理を実行する対象の入力データ。
+
+        Returns
+        -------
+        result : dict
+            推論処理の結果を保持する辞書型データ。
+            基本的にinput_dataと同じ構造です。
+        """
+        result = []
+        print('### Line OCR Process ###')
+        result_xml = self._run_src_inference(input_data['img'], input_data['xml'],
+                                             accept_empty=self.cfg['line_ocr']['accept_empty'],
+                                             yield_block_page_num=self.cfg['line_ocr']['yield_block_page_num'],
+                                             yield_block_pillar=self.cfg['line_ocr']['yield_block_pillar'])
+
+        output_data = copy.deepcopy(input_data)
+        output_data['xml'] = result_xml
+        result.append(output_data)
+
+        return result

cli/procs/page_deskew.py

@@ -0,0 +1,85 @@
+# Copyright (c) 2022, National Diet Library, Japan
+#
+# This software is released under the CC BY 4.0.
+# https://creativecommons.org/licenses/by/4.0/
+
+
+import copy
+import numpy
+
+from .base_proc import BaseInferenceProcess
+
+
+class PageDeskewProcess(BaseInferenceProcess):
+    """
+    傾き補正を実行するプロセスのクラス。
+    BaseInferenceProcessを継承しています。
+    """
+    def __init__(self, cfg, pid):
+        """
+        Parameters
+        ----------
+        cfg : dict
+            本推論処理における設定情報です。
+        pid : int
+            実行される順序を表す数値。
+        """
+        super().__init__(cfg, pid, '_page_deskew')
+        from src.deskew_HT.alyn3.deskew import Deskew
+        self.deskewer = Deskew('', '',
+                               r_angle=cfg['page_deskew']['r_angle'],
+                               skew_max=cfg['page_deskew']['skew_max'],
+                               acc_deg=cfg['page_deskew']['acc_deg'],
+                               method=cfg['page_deskew']['method'],
+                               gray=cfg['page_deskew']['gray'],
+                               quality=cfg['page_deskew']['quality'],
+                               short=cfg['page_deskew']['short'],
+                               roi_w=cfg['page_deskew']['roi_w'],
+                               roi_h=cfg['page_deskew']['roi_h'])
+        self._run_src_inference = self.deskewer.deskew_on_memory
+
+
+    def _is_valid_input(self, input_data):
+        """
+        本クラスの推論処理における入力データのバリデーション。
+
+        Parameters
+        ----------
+        input_data : dict
+            推論処理を実行する対象の入力データ。
+
+        Returns
+        -------
+        [変数なし] : bool
+            　入力データが正しければTrue, そうでなければFalseを返します。
+        """
+        if type(input_data['img']) is not numpy.ndarray:
+            print('PageDeskewProcess: input img is not numpy.ndarray')
+            return False
+        return True
+
+    def _run_process(self, input_data):
+        """
+        推論処理の本体部分。
+
+        Parameters
+        ----------
+        input_data : dict
+            推論処理を実行する対象の入力データ。
+
+        Returns
+        -------
+        result : dict
+            推論処理の結果を保持する辞書型データ。
+            基本的にinput_dataと同じ構造です。
+        """
+        print('### Page Deskew Process ###')
+        inference_output = self._run_src_inference(input_data['img'])
+
+        # Create result to pass img_path and img data
+        result = []
+        output_data = copy.deepcopy(input_data)
+        output_data['img'] = inference_output
+        result.append(output_data)
+
+        return result

cli/procs/page_separation.py

@@ -0,0 +1,112 @@
+# Copyright (c) 2022, National Diet Library, Japan
+#
+# This software is released under the CC BY 4.0.
+# https://creativecommons.org/licenses/by/4.0/
+
+
+import copy
+import numpy
+import os
+
+from .base_proc import BaseInferenceProcess
+
+
+class PageSeparation(BaseInferenceProcess):
+    """
+    ノド元分割処理を実行するプロセスのクラス。
+    BaseInferenceProcessを継承しています。
+    """
+    def __init__(self, cfg, pid):
+        """
+        Parameters
+        ----------
+        cfg : dict
+            本推論処理における設定情報です。
+        pid : int
+            実行される順序を表す数値。
+        """
+        super().__init__(cfg, pid, '_page_sep')
+
+        if self.cfg['page_separation']['silence_tf_log']:
+            import logging
+            import warnings
+            os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'
+            warnings.simplefilter(action='ignore', category=FutureWarning)
+
+            import tensorflow as tf
+            tf.get_logger().setLevel(logging.ERROR)
+
+        from src.separate_pages_ssd.inference_divided import divide_facing_page_with_cli, load_weightfile
+        load_weightfile(os.path.abspath(self.cfg['page_separation']['weight_path']))
+        self._run_src_inference = divide_facing_page_with_cli
+
+    def _is_valid_input(self, input_data):
+        """
+        本クラスの推論処理における入力データのバリデーション。
+
+        Parameters
+        ----------
+        input_data : dict
+            推論処理を実行する対象の入力データ。
+
+        Returns
+        -------
+        [変数なし] : bool
+            　入力データが正しければTrue, そうでなければFalseを返します。
+        """
+        if type(input_data['img']) is not numpy.ndarray:
+            print('PageSeparation: input img is not numpy.ndarray')
+            return False
+        return True
+
+    def _run_process(self, input_data):
+        """
+        推論処理の本体部分。
+
+        Parameters
+        ----------
+        input_data : dict
+            推論処理を実行する対象の入力データ。
+
+        Returns
+        -------
+        result : dict
+            推論処理の結果を保持する辞書型データ。
+            基本的にinput_dataと同じ構造です。
+        """
+        print('### Page Separation ###')
+        log_file_path = None
+        if self.process_dump_dir is not None:
+            log_file_path = os.path.join(self.process_dump_dir, self.cfg['page_separation']['log'])
+        inference_output = self._run_src_inference(input=input_data['img'],
+                                                   input_path=input_data['img_path'],
+                                                   left=self.cfg['page_separation']['left'],
+                                                   right=self.cfg['page_separation']['right'],
+                                                   single=self.cfg['page_separation']['single'],
+                                                   ext=self.cfg['page_separation']['ext'],
+                                                   quality=self.cfg['page_separation']['quality'],
+                                                   short=self.cfg['page_separation']['short'],
+                                                   log=log_file_path)
+        if (not self.cfg['page_separation']['allow_invalid_num_output']) and (not len(inference_output) in range(1, 3)):
+            print('ERROR: Output from page separation must be 1 or 2 pages.')
+            return None
+
+        # Create result to pass img_path and img data
+        result = []
+        for id, single_output_img in enumerate(inference_output):
+            output_data = copy.deepcopy(input_data)
+            output_data['img'] = single_output_img
+            output_data['orig_img_path'] = input_data['img_path']
+
+            # make and save separated img file name
+            if id == 0:
+                id = 'L'
+            else:
+                id = 'R'
+            orig_img_name = os.path.basename(input_data['img_path'])
+            stem, ext = os.path.splitext(orig_img_name)
+            output_data['img_file_name'] = stem + '_' + id + '.jpg'
+
+            result.append(output_data)
+
+        return result

config.yml

@@ -0,0 +1,45 @@
+page_separation:
+  input: 'inference_input'
+  out: 'NO_DUMP'
+  left: '_01'
+  right: '_02'
+  single: '_00'
+  ext: '.jpg'
+  quality: 100
+  short: null
+  log: 'trim_pos.tsv'
+  weight_path: 'src/separate_pages_ssd/ssd_tools/weights.hdf5'
+  allow_invalid_num_output: True 
+  silence_tf_log: True
+page_deskew:
+  r_angle: 0
+  skew_max: 4.0
+  acc_deg: 0.1
+  method: 1
+  gray: 1.0
+  quality: 100
+  short: null
+  roi_w: 1.0
+  roi_h: 1.0
+layout_extraction:
+  config_path: 'src/ndl_layout/models/ndl_layout_config.py'
+  checkpoint_path: 'src/ndl_layout/models/epoch_140_all_eql_bt.pth'
+  device: 'cpu'
+  'score_thr': 0.3
+line_ocr:
+  char_list: 'src/text_recognition/models/mojilist_NDL.txt'
+  saved_model: 'src/text_recognition/models/ndlenfixed64-mj0-synth1.pth'
+  accept_empty: True
+  batch_max_length: 100
+  batch_size: 100
+  character: None
+  imgW: 1200
+  workers: 1
+  xml: None
+  yield_block_pillar: True
+  yield_block_page_num: True
+  FeatureExtraction: 'ResNet'
+  Prediction: 'CTC'
+  PAD: True
+  SequenceModeling: 'None'
+  Transformation: 'None'

docker/Dockerfile

@@ -0,0 +1,36 @@
+FROM nvcr.io/nvidia/cuda:11.1.1-cudnn8-devel-ubuntu18.04
+
+ENV PROJECT_DIR=/root/ocr_cli
+ENV FORCE_CUDA="1"
+ENV TORCH_CUDA_ARCH_LIST="7.5+PTX"
+ENV TORCH_NVCC_FLAGS="-Xfatbin -compress-all"
+
+RUN set -x \
+    && apt update \
+    && apt upgrade -y
+
+RUN set -x \
+    && apt update \
+    && apt -y install locales \
+    && locale-gen ja_JP.UTF-8
+ENV LANG ja_JP.UTF-8
+ENV LANGUAGE ja_JP:ja
+ENV LC_ALL=ja_JP.UTF-8
+RUN localedef -f UTF-8 -i ja_JP ja_JP.utf8
+
+RUN set -x && apt -y install libgl1-mesa-dev libglib2.0-0 git
+RUN set -x \
+    && apt -y install python3.7 python3.7-dev \
+    && ln -s /usr/bin/python3.7 /usr/bin/python \
+    && apt -y install wget python3-distutils && wget https://bootstrap.pypa.io/get-pip.py && python get-pip.py
+
+COPY . ${PROJECT_DIR}
+
+RUN set -x \
+    && pip install -r ${PROJECT_DIR}/requirements.txt
+RUN set -x && pip install torch==1.8.1+cu111 torchvision==0.9.1+cu111 -f https://download.pytorch.org/whl/lts/1.8/torch_lts.html
+RUN set -x && cd ${PROJECT_DIR}/src/ndl_layout/mmdetection && python setup.py bdist_wheel && pip install dist/*.whl
+ENV PYTHONPATH $PYTHONPATH:${PROJECT_DIR}/src/text_recognition/deep-text-recognition-benchmark
+RUN set -x && pip install mmcv-full==1.4.0 -f https://download.openmmlab.com/mmcv/dist/cu111/torch1.8.0/index.html
+
+WORKDIR ${PROJECT_DIR}

docker/dockerbuild.sh

@@ -0,0 +1,12 @@
+TAG=ocr-cli-py37
+DOCKERIGNORE=docker/dockerignore
+DOCKERFILE=docker/Dockerfile
+wget https://lab.ndl.go.jp/dataset/ndlocr/text_recognition/mojilist_NDL.txt -P ./src/text_recognition/models
+wget https://lab.ndl.go.jp/dataset/ndlocr/text_recognition/ndlenfixed64-mj0-synth1.pth -P ./src/text_recognition/models
+wget https://lab.ndl.go.jp/dataset/ndlocr/ndl_layout/ndl_layout_config.py -P ./src/ndl_layout/models
+wget https://lab.ndl.go.jp/dataset/ndlocr/ndl_layout/epoch_140_all_eql_bt.pth -P ./src/ndl_layout/models
+wget https://lab.ndl.go.jp/dataset/ndlocr/separate_pages_ssd/weights.hdf5 -P ./src/separate_pages_ssd/ssd_tools
+
+cp ${DOCKERIGNORE} .dockerignore
+docker build -t ${TAG} -f ${DOCKERFILE} .
+rm .dockerignore

docker/dockerignore

@@ -0,0 +1,4 @@
+**/*.pyc
+**/*.pyd
+**/*~
+**/.*

docker/run_docker.sh

@@ -0,0 +1 @@
+docker run --gpus all -d --rm --name ocr_cli_runner -i ocr-cli-py37:latest

main.py

@@ -0,0 +1,92 @@
+# Copyright (c) 2022, National Diet Library, Japan
+#
+# This software is released under the CC BY 4.0.
+# https://creativecommons.org/licenses/by/4.0/
+
+
+import click
+import json
+import os
+import sys
+
+from cli.core import OcrInferencer
+from cli.core import utils
+
+
+@click.group()
+@click.option('--debug', is_flag=True)
+@click.pass_context
+def cmd(ctx, debug):
+    ctx.obj['DEBUG'] = debug
+
+
+@cmd.command()
+@click.pass_context
+def help(ctx):
+    if ctx.obj['DEBUG']:
+        click.echo('DEBUG MODE!')
+    click.echo('help!')
+
+
+@cmd.command()
+@click.pass_context
+@click.argument('input_root')
+@click.argument('output_root')
+@click.option('-s', '--input_structure', type=click.Choice(['s', 'i', 't', 'w', 'f'], case_sensitive=True), default='s', help='Input directory structure type. s(single), i(intermediate_output), t(tosho_data), w(workstation), and f(image_file).')
+@click.option('-p', '--proc_range', type=str, default='0..3', help='Inference process range to run. Default is "0..3".')
+@click.option('-c', '--config_file', type=str, default='config.yml', help='Configuration yml file for inference. Default is "config.yml".')
+@click.option('-i', '--save_image', type=bool, default=False, is_flag=True, help='Output result image file with text file.')
+@click.option('-x', '--save_xml', type=bool, default=False, is_flag=True, help='Output result XML file with text file.')
+@click.option('-d', '--dump', type=bool, default=False, is_flag=True, help='Dump all intermediate process output.')
+def infer(ctx, input_root, output_root, config_file, proc_range, save_image, save_xml, input_structure, dump):
+    """
+    \b
+    INPUT_ROOT   \t: Input data directory for inference.
+    OUTPUT_ROOT   \t: Output directory for inference result.
+    """
+    click.echo('start inference !')
+    click.echo('input_root : {0}'.format(input_root))
+    click.echo('output_root : {0}'.format(output_root))
+    click.echo('config_file : {0}'.format(config_file))
+
+    cfg = {
+        'input_root': input_root,
+        'output_root': output_root,
+        'config_file': config_file,
+        'proc_range': proc_range,
+        'save_image': save_image,
+        'save_xml': save_xml,
+        'dump': dump,
+        'input_structure': input_structure
+    }
+
+    # check if input_root exists
+    if not os.path.exists(input_root):
+        print('INPUT_ROOT not found :{0}'.format(input_root), file=sys.stderr)
+        exit(0)
+
+    # parse command line option
+    infer_cfg = utils.parse_cfg(cfg)
+    if infer_cfg is None:
+        print('[ERROR] Config parse error :{0}'.format(input_root), file=sys.stderr)
+        exit(1)
+
+    # prepare output root derectory
+    infer_cfg['output_root'] = utils.mkdir_with_duplication_check(infer_cfg['output_root'])
+
+    # save inference option
+    with open(os.path.join(infer_cfg['output_root'], 'opt.json'), 'w') as fp:
+        json.dump(infer_cfg, fp, ensure_ascii=False, indent=4,
+                  sort_keys=True, separators=(',', ': '))
+
+    # do inference
+    inferencer = OcrInferencer(infer_cfg)
+    inferencer.run()
+
+
+def main():
+    cmd(obj={})
+
+
+if __name__ == '__main__':
+    main()

requirements.txt

@@ -0,0 +1,13 @@
+click
+keras==2.2.4
+lmdb==1.2.1
+natsort==7.1.1
+nltk==3.6.2
+opencv-python==4.5.1.48
+pyyaml
+scikit-image==0.16.2
+scipy==1.1.0
+tensorboard==1.14.0
+tensorflow==1.14.0
+tensorflow-estimator==1.14.0
+wheel

src/deskew_HT/.gitignore

@@ -0,0 +1,16 @@
+# image files
+*.jp2
+*.jpg
+*.JPG
+*.png
+
+# log
+*.txt
+
+# annotation
+*.xml
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class

src/deskew_HT/LICENSE

@@ -0,0 +1,432 @@
+alyn3 is based on the following repository:
+
+https://github.com/kakul/Alyn/tree/master/alyn
+
+The original software license is as follows.
+
+-----------------------------------------------------------------------
+
+The MIT License (MIT)
+
+Copyright (c) 2016 Kakul
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+
+-----------------------------------------------------------------------
+
+The following license applies for other modifications in this repository:
+
+-----------------------------------------------------------------------
+
+
+Copyright (c) 2022, National Diet Library, Japan.
+
+Attribution 4.0 International
+
+=======================================================================
+
+Creative Commons Corporation ("Creative Commons") is not a law firm and
+does not provide legal services or legal advice. Distribution of
+Creative Commons public licenses does not create a lawyer-client or
+other relationship. Creative Commons makes its licenses and related
+information available on an "as-is" basis. Creative Commons gives no
+warranties regarding its licenses, any material licensed under their
+terms and conditions, or any related information. Creative Commons
+disclaims all liability for damages resulting from their use to the
+fullest extent possible.
+
+Using Creative Commons Public Licenses
+
+Creative Commons public licenses provide a standard set of terms and
+conditions that creators and other rights holders may use to share
+original works of authorship and other material subject to copyright
+and certain other rights specified in the public license below. The
+following considerations are for informational purposes only, are not
+exhaustive, and do not form part of our licenses.
+
+     Considerations for licensors: Our public licenses are
+     intended for use by those authorized to give the public
+     permission to use material in ways otherwise restricted by
+     copyright and certain other rights. Our licenses are
+     irrevocable. Licensors should read and understand the terms
+     and conditions of the license they choose before applying it.
+     Licensors should also secure all rights necessary before
+     applying our licenses so that the public can reuse the
+     material as expected. Licensors should clearly mark any
+     material not subject to the license. This includes other CC-
+     licensed material, or material used under an exception or
+     limitation to copyright. More considerations for licensors:
+	wiki.creativecommons.org/Considerations_for_licensors
+
+     Considerations for the public: By using one of our public
+     licenses, a licensor grants the public permission to use the
+     licensed material under specified terms and conditions. If
+     the licensor's permission is not necessary for any reason--for
+     example, because of any applicable exception or limitation to
+     copyright--then that use is not regulated by the license. Our
+     licenses grant only permissions under copyright and certain
+     other rights that a licensor has authority to grant. Use of
+     the licensed material may still be restricted for other
+     reasons, including because others have copyright or other
+     rights in the material. A licensor may make special requests,
+     such as asking that all changes be marked or described.
+     Although not required by our licenses, you are encouraged to
+     respect those requests where reasonable. More_considerations
+     for the public: 
+	wiki.creativecommons.org/Considerations_for_licensees
+
+=======================================================================
+
+Creative Commons Attribution 4.0 International Public License
+
+By exercising the Licensed Rights (defined below), You accept and agree
+to be bound by the terms and conditions of this Creative Commons
+Attribution 4.0 International Public License ("Public License"). To the
+extent this Public License may be interpreted as a contract, You are
+granted the Licensed Rights in consideration of Your acceptance of
+these terms and conditions, and the Licensor grants You such rights in
+consideration of benefits the Licensor receives from making the
+Licensed Material available under these terms and conditions.
+
+
+Section 1 -- Definitions.
+
+  a. Adapted Material means material subject to Copyright and Similar
+     Rights that is derived from or based upon the Licensed Material
+     and in which the Licensed Material is translated, altered,
+     arranged, transformed, or otherwise modified in a manner requiring
+     permission under the Copyright and Similar Rights held by the
+     Licensor. For purposes of this Public License, where the Licensed
+     Material is a musical work, performance, or sound recording,
+     Adapted Material is always produced where the Licensed Material is
+     synched in timed relation with a moving image.
+
+  b. Adapter's License means the license You apply to Your Copyright
+     and Similar Rights in Your contributions to Adapted Material in
+     accordance with the terms and conditions of this Public License.
+
+  c. Copyright and Similar Rights means copyright and/or similar rights
+     closely related to copyright including, without limitation,
+     performance, broadcast, sound recording, and Sui Generis Database
+     Rights, without regard to how the rights are labeled or
+     categorized. For purposes of this Public License, the rights
+     specified in Section 2(b)(1)-(2) are not Copyright and Similar
+     Rights.
+
+  d. Effective Technological Measures means those measures that, in the
+     absence of proper authority, may not be circumvented under laws
+     fulfilling obligations under Article 11 of the WIPO Copyright
+     Treaty adopted on December 20, 1996, and/or similar international
+     agreements.
+
+  e. Exceptions and Limitations means fair use, fair dealing, and/or
+     any other exception or limitation to Copyright and Similar Rights
+     that applies to Your use of the Licensed Material.
+
+  f. Licensed Material means the artistic or literary work, database,
+     or other material to which the Licensor applied this Public
+     License.
+
+  g. Licensed Rights means the rights granted to You subject to the
+     terms and conditions of this Public License, which are limited to
+     all Copyright and Similar Rights that apply to Your use of the
+     Licensed Material and that the Licensor has authority to license.
+
+  h. Licensor means the individual(s) or entity(ies) granting rights
+     under this Public License.
+
+  i. Share means to provide material to the public by any means or
+     process that requires permission under the Licensed Rights, such
+     as reproduction, public display, public performance, distribution,
+     dissemination, communication, or importation, and to make material
+     available to the public including in ways that members of the
+     public may access the material from a place and at a time
+     individually chosen by them.
+
+  j. Sui Generis Database Rights means rights other than copyright
+     resulting from Directive 96/9/EC of the European Parliament and of
+     the Council of 11 March 1996 on the legal protection of databases,
+     as amended and/or succeeded, as well as other essentially
+     equivalent rights anywhere in the world.
+
+  k. You means the individual or entity exercising the Licensed Rights
+     under this Public License. Your has a corresponding meaning.
+
+
+Section 2 -- Scope.
+
+  a. License grant.
+
+       1. Subject to the terms and conditions of this Public License,
+          the Licensor hereby grants You a worldwide, royalty-free,
+          non-sublicensable, non-exclusive, irrevocable license to
+          exercise the Licensed Rights in the Licensed Material to:
+
+            a. reproduce and Share the Licensed Material, in whole or
+               in part; and
+
+            b. produce, reproduce, and Share Adapted Material.
+
+       2. Exceptions and Limitations. For the avoidance of doubt, where
+          Exceptions and Limitations apply to Your use, this Public
+          License does not apply, and You do not need to comply with
+          its terms and conditions.
+
+       3. Term. The term of this Public License is specified in Section
+          6(a).
+
+       4. Media and formats; technical modifications allowed. The
+          Licensor authorizes You to exercise the Licensed Rights in
+          all media and formats whether now known or hereafter created,
+          and to make technical modifications necessary to do so. The
+          Licensor waives and/or agrees not to assert any right or
+          authority to forbid You from making technical modifications
+          necessary to exercise the Licensed Rights, including
+          technical modifications necessary to circumvent Effective
+          Technological Measures. For purposes of this Public License,
+          simply making modifications authorized by this Section 2(a)
+          (4) never produces Adapted Material.
+
+       5. Downstream recipients.
+
+            a. Offer from the Licensor -- Licensed Material. Every
+               recipient of the Licensed Material automatically
+               receives an offer from the Licensor to exercise the
+               Licensed Rights under the terms and conditions of this
+               Public License.
+
+            b. No downstream restrictions. You may not offer or impose
+               any additional or different terms or conditions on, or
+               apply any Effective Technological Measures to, the
+               Licensed Material if doing so restricts exercise of the
+               Licensed Rights by any recipient of the Licensed
+               Material.
+
+       6. No endorsement. Nothing in this Public License constitutes or
+          may be construed as permission to assert or imply that You
+          are, or that Your use of the Licensed Material is, connected
+          with, or sponsored, endorsed, or granted official status by,
+          the Licensor or others designated to receive attribution as
+          provided in Section 3(a)(1)(A)(i).
+
+  b. Other rights.
+
+       1. Moral rights, such as the right of integrity, are not
+          licensed under this Public License, nor are publicity,
+          privacy, and/or other similar personality rights; however, to
+          the extent possible, the Licensor waives and/or agrees not to
+          assert any such rights held by the Licensor to the limited
+          extent necessary to allow You to exercise the Licensed
+          Rights, but not otherwise.
+
+       2. Patent and trademark rights are not licensed under this
+          Public License.
+
+       3. To the extent possible, the Licensor waives any right to
+          collect royalties from You for the exercise of the Licensed
+          Rights, whether directly or through a collecting society
+          under any voluntary or waivable statutory or compulsory
+          licensing scheme. In all other cases the Licensor expressly
+          reserves any right to collect such royalties.
+
+
+Section 3 -- License Conditions.
+
+Your exercise of the Licensed Rights is expressly made subject to the
+following conditions.
+
+  a. Attribution.
+
+       1. If You Share the Licensed Material (including in modified
+          form), You must:
+
+            a. retain the following if it is supplied by the Licensor
+               with the Licensed Material:
+
+                 i. identification of the creator(s) of the Licensed
+                    Material and any others designated to receive
+                    attribution, in any reasonable manner requested by
+                    the Licensor (including by pseudonym if
+                    designated);
+
+                ii. a copyright notice;
+
+               iii. a notice that refers to this Public License;
+
+                iv. a notice that refers to the disclaimer of
+                    warranties;
+
+                 v. a URI or hyperlink to the Licensed Material to the
+                    extent reasonably practicable;
+
+            b. indicate if You modified the Licensed Material and
+               retain an indication of any previous modifications; and
+
+            c. indicate the Licensed Material is licensed under this
+               Public License, and include the text of, or the URI or
+               hyperlink to, this Public License.
+
+       2. You may satisfy the conditions in Section 3(a)(1) in any
+          reasonable manner based on the medium, means, and context in
+          which You Share the Licensed Material. For example, it may be
+          reasonable to satisfy the conditions by providing a URI or
+          hyperlink to a resource that includes the required
+          information.
+
+       3. If requested by the Licensor, You must remove any of the
+          information required by Section 3(a)(1)(A) to the extent
+          reasonably practicable.
+
+       4. If You Share Adapted Material You produce, the Adapter's
+          License You apply must not prevent recipients of the Adapted
+          Material from complying with this Public License.
+
+
+Section 4 -- Sui Generis Database Rights.
+
+Where the Licensed Rights include Sui Generis Database Rights that
+apply to Your use of the Licensed Material:
+
+  a. for the avoidance of doubt, Section 2(a)(1) grants You the right
+     to extract, reuse, reproduce, and Share all or a substantial
+     portion of the contents of the database;
+
+  b. if You include all or a substantial portion of the database
+     contents in a database in which You have Sui Generis Database
+     Rights, then the database in which You have Sui Generis Database
+     Rights (but not its individual contents) is Adapted Material; and
+
+  c. You must comply with the conditions in Section 3(a) if You Share
+     all or a substantial portion of the contents of the database.
+
+For the avoidance of doubt, this Section 4 supplements and does not
+replace Your obligations under this Public License where the Licensed
+Rights include other Copyright and Similar Rights.
+
+
+Section 5 -- Disclaimer of Warranties and Limitation of Liability.
+
+  a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
+     EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
+     AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
+     ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
+     IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
+     WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
+     PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
+     ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
+     KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
+     ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
+
+  b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
+     TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
+     NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
+     INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
+     COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
+     USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
+     ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
+     DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
+     IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
+
+  c. The disclaimer of warranties and limitation of liability provided
+     above shall be interpreted in a manner that, to the extent
+     possible, most closely approximates an absolute disclaimer and
+     waiver of all liability.
+
+
+Section 6 -- Term and Termination.
+
+  a. This Public License applies for the term of the Copyright and
+     Similar Rights licensed here. However, if You fail to comply with
+     this Public License, then Your rights under this Public License
+     terminate automatically.
+
+  b. Where Your right to use the Licensed Material has terminated under
+     Section 6(a), it reinstates:
+
+       1. automatically as of the date the violation is cured, provided
+          it is cured within 30 days of Your discovery of the
+          violation; or
+
+       2. upon express reinstatement by the Licensor.
+
+     For the avoidance of doubt, this Section 6(b) does not affect any
+     right the Licensor may have to seek remedies for Your violations
+     of this Public License.
+
+  c. For the avoidance of doubt, the Licensor may also offer the
+     Licensed Material under separate terms or conditions or stop
+     distributing the Licensed Material at any time; however, doing so
+     will not terminate this Public License.
+
+  d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
+     License.
+
+
+Section 7 -- Other Terms and Conditions.
+
+  a. The Licensor shall not be bound by any additional or different
+     terms or conditions communicated by You unless expressly agreed.
+
+  b. Any arrangements, understandings, or agreements regarding the
+     Licensed Material not stated herein are separate from and
+     independent of the terms and conditions of this Public License.
+
+
+Section 8 -- Interpretation.
+
+  a. For the avoidance of doubt, this Public License does not, and
+     shall not be interpreted to, reduce, limit, restrict, or impose
+     conditions on any use of the Licensed Material that could lawfully
+     be made without permission under this Public License.
+
+  b. To the extent possible, if any provision of this Public License is
+     deemed unenforceable, it shall be automatically reformed to the
+     minimum extent necessary to make it enforceable. If the provision
+     cannot be reformed, it shall be severed from this Public License
+     without affecting the enforceability of the remaining terms and
+     conditions.
+
+  c. No term or condition of this Public License will be waived and no
+     failure to comply consented to unless expressly agreed to by the
+     Licensor.
+
+  d. Nothing in this Public License constitutes or may be interpreted
+     as a limitation upon, or waiver of, any privileges and immunities
+     that apply to the Licensor or You, including from the legal
+     processes of any jurisdiction or authority.
+
+
+=======================================================================
+
+Creative Commons is not a party to its public licenses.
+Notwithstanding, Creative Commons may elect to apply one of its public
+licenses to material it publishes and in those instances will be
+considered the "Licensor." Except for the limited purpose of indicating
+that material is shared under a Creative Commons public license or as
+otherwise permitted by the Creative Commons policies published at
+creativecommons.org/policies, Creative Commons does not authorize the
+use of the trademark "Creative Commons" or any other trademark or logo
+of Creative Commons without its prior written consent including,
+without limitation, in connection with any unauthorized modifications
+to any of its public licenses or any other arrangements,
+understandings, or agreements concerning use of licensed material. For
+the avoidance of doubt, this paragraph does not form part of the public
+licenses.
+
+Creative Commons may be contacted at creativecommons.org.

src/deskew_HT/LICENSE_DEPENDENCIES

@@ -0,0 +1,677 @@
+# Pillow
+
+The Python Imaging Library (PIL) is
+
+    Copyright © 1997-2011 by Secret Labs AB
+    Copyright © 1995-2011 by Fredrik Lundh
+
+Pillow is the friendly PIL fork. It is
+
+    Copyright © 2010-2022 by Alex Clark and contributors
+
+Like PIL, Pillow is licensed under the open source HPND License:
+
+By obtaining, using, and/or copying this software and/or its associated
+documentation, you agree that you have read, understood, and will comply
+with the following terms and conditions:
+
+Permission to use, copy, modify, and distribute this software and its
+associated documentation for any purpose and without fee is hereby granted,
+provided that the above copyright notice appears in all copies, and that
+both that copyright notice and this permission notice appear in supporting
+documentation, and that the name of Secret Labs AB or the author not be
+used in advertising or publicity pertaining to distribution of the software
+without specific, written prior permission.
+
+SECRET LABS AB AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS
+SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS.
+IN NO EVENT SHALL SECRET LABS AB OR THE AUTHOR BE LIABLE FOR ANY SPECIAL,
+INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
+LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE
+OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
+PERFORMANCE OF THIS SOFTWARE.
+
+
+# PyWavelets
+
+Copyright (c) 2006-2012 Filip Wasilewski <http://en.ig.ma/>
+Copyright (c) 2012-2020 The PyWavelets Developers <https://github.com/PyWavelets/pywt>
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of
+this software and associated documentation files (the "Software"), to deal in
+the Software without restriction, including without limitation the rights to
+use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+of the Software, and to permit persons to whom the Software is furnished to do
+so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+
+
+# cycler
+
+Copyright (c) 2015, matplotlib project
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+* Redistributions of source code must retain the above copyright notice, this
+  list of conditions and the following disclaimer.
+
+* Redistributions in binary form must reproduce the above copyright notice,
+  this list of conditions and the following disclaimer in the documentation
+  and/or other materials provided with the distribution.
+
+* Neither the name of the matplotlib project nor the names of its
+  contributors may be used to endorse or promote products derived from
+  this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+# decorator
+
+Copyright (c) 2005-2018, Michele Simionato
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+  Redistributions of source code must retain the above copyright
+  notice, this list of conditions and the following disclaimer.
+  Redistributions in bytecode form must reproduce the above copyright
+  notice, this list of conditions and the following disclaimer in
+  the documentation and/or other materials provided with the
+  distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
+INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
+OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
+TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
+USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
+DAMAGE.
+
+
+# imageio
+
+Copyright (c) 2014-2020, imageio developers
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+* Redistributions of source code must retain the above copyright notice, this
+  list of conditions and the following disclaimer.
+
+* Redistributions in binary form must reproduce the above copyright notice,
+  this list of conditions and the following disclaimer in the documentation
+  and/or other materials provided with the distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+# kiwisolver
+
+=========================
+ The Kiwi licensing terms
+=========================
+Kiwi is licensed under the terms of the Modified BSD License (also known as
+New or Revised BSD), as follows:
+
+Copyright (c) 2013, Nucleic Development Team
+
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+Redistributions of source code must retain the above copyright notice, this
+list of conditions and the following disclaimer.
+
+Redistributions in binary form must reproduce the above copyright notice, this
+list of conditions and the following disclaimer in the documentation and/or
+other materials provided with the distribution.
+
+Neither the name of the Nucleic Development Team nor the names of its
+contributors may be used to endorse or promote products derived from this
+software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+About Kiwi
+----------
+Chris Colbert began the Kiwi project in December 2013 in an effort to
+create a blisteringly fast UI constraint solver. Chris is still the
+project lead.
+
+The Nucleic Development Team is the set of all contributors to the Nucleic
+project and its subprojects.
+
+The core team that coordinates development on GitHub can be found here:
+http://github.com/nucleic. The current team consists of:
+
+* Chris Colbert
+
+Our Copyright Policy
+--------------------
+Nucleic uses a shared copyright model. Each contributor maintains copyright
+over their contributions to Nucleic. But, it is important to note that these
+contributions are typically only changes to the repositories. Thus, the Nucleic
+source code, in its entirety is not the copyright of any single person or
+institution. Instead, it is the collective copyright of the entire Nucleic
+Development Team. If individual contributors want to maintain a record of what
+changes/contributions they have specific copyright on, they should indicate
+their copyright in the commit message of the change, when they commit the
+change to one of the Nucleic repositories.
+
+With this in mind, the following banner should be used in any source code file
+to indicate the copyright and license terms:
+
+#------------------------------------------------------------------------------
+# Copyright (c) 2013, Nucleic Development Team.
+#
+# Distributed under the terms of the Modified BSD License.
+#
+# The full license is in the file LICENSE, distributed with this software.
+#------------------------------------------------------------------------------
+
+
+# matplotlib
+
+License agreement for matplotlib versions 1.3.0 and later
+=========================================================
+
+1. This LICENSE AGREEMENT is between the Matplotlib Development Team
+("MDT"), and the Individual or Organization ("Licensee") accessing and
+otherwise using matplotlib software in source or binary form and its
+associated documentation.
+
+2. Subject to the terms and conditions of this License Agreement, MDT
+hereby grants Licensee a nonexclusive, royalty-free, world-wide license
+to reproduce, analyze, test, perform and/or display publicly, prepare
+derivative works, distribute, and otherwise use matplotlib
+alone or in any derivative version, provided, however, that MDT's
+License Agreement and MDT's notice of copyright, i.e., "Copyright (c)
+2012- Matplotlib Development Team; All Rights Reserved" are retained in
+matplotlib  alone or in any derivative version prepared by
+Licensee.
+
+3. In the event Licensee prepares a derivative work that is based on or
+incorporates matplotlib or any part thereof, and wants to
+make the derivative work available to others as provided herein, then
+Licensee hereby agrees to include in any such work a brief summary of
+the changes made to matplotlib .
+
+4. MDT is making matplotlib available to Licensee on an "AS
+IS" basis.  MDT MAKES NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR
+IMPLIED.  BY WAY OF EXAMPLE, BUT NOT LIMITATION, MDT MAKES NO AND
+DISCLAIMS ANY REPRESENTATION OR WARRANTY OF MERCHANTABILITY OR FITNESS
+FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF MATPLOTLIB
+WILL NOT INFRINGE ANY THIRD PARTY RIGHTS.
+
+5. MDT SHALL NOT BE LIABLE TO LICENSEE OR ANY OTHER USERS OF MATPLOTLIB
+ FOR ANY INCIDENTAL, SPECIAL, OR CONSEQUENTIAL DAMAGES OR
+LOSS AS A RESULT OF MODIFYING, DISTRIBUTING, OR OTHERWISE USING
+MATPLOTLIB , OR ANY DERIVATIVE THEREOF, EVEN IF ADVISED OF
+THE POSSIBILITY THEREOF.
+
+6. This License Agreement will automatically terminate upon a material
+breach of its terms and conditions.
+
+7. Nothing in this License Agreement shall be deemed to create any
+relationship of agency, partnership, or joint venture between MDT and
+Licensee.  This License Agreement does not grant permission to use MDT
+trademarks or trade name in a trademark sense to endorse or promote
+products or services of Licensee, or any third party.
+
+8. By copying, installing or otherwise using matplotlib ,
+Licensee agrees to be bound by the terms and conditions of this License
+Agreement.
+
+License agreement for matplotlib versions prior to 1.3.0
+========================================================
+
+1. This LICENSE AGREEMENT is between John D. Hunter ("JDH"), and the
+Individual or Organization ("Licensee") accessing and otherwise using
+matplotlib software in source or binary form and its associated
+documentation.
+
+2. Subject to the terms and conditions of this License Agreement, JDH
+hereby grants Licensee a nonexclusive, royalty-free, world-wide license
+to reproduce, analyze, test, perform and/or display publicly, prepare
+derivative works, distribute, and otherwise use matplotlib
+alone or in any derivative version, provided, however, that JDH's
+License Agreement and JDH's notice of copyright, i.e., "Copyright (c)
+2002-2011 John D. Hunter; All Rights Reserved" are retained in
+matplotlib  alone or in any derivative version prepared by
+Licensee.
+
+3. In the event Licensee prepares a derivative work that is based on or
+incorporates matplotlib  or any part thereof, and wants to
+make the derivative work available to others as provided herein, then
+Licensee hereby agrees to include in any such work a brief summary of
+the changes made to matplotlib.
+
+4. JDH is making matplotlib  available to Licensee on an "AS
+IS" basis.  JDH MAKES NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR
+IMPLIED.  BY WAY OF EXAMPLE, BUT NOT LIMITATION, JDH MAKES NO AND
+DISCLAIMS ANY REPRESENTATION OR WARRANTY OF MERCHANTABILITY OR FITNESS
+FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF MATPLOTLIB
+WILL NOT INFRINGE ANY THIRD PARTY RIGHTS.
+
+5. JDH SHALL NOT BE LIABLE TO LICENSEE OR ANY OTHER USERS OF MATPLOTLIB
+ FOR ANY INCIDENTAL, SPECIAL, OR CONSEQUENTIAL DAMAGES OR
+LOSS AS A RESULT OF MODIFYING, DISTRIBUTING, OR OTHERWISE USING
+MATPLOTLIB , OR ANY DERIVATIVE THEREOF, EVEN IF ADVISED OF
+THE POSSIBILITY THEREOF.
+
+6. This License Agreement will automatically terminate upon a material
+breach of its terms and conditions.
+
+7. Nothing in this License Agreement shall be deemed to create any
+relationship of agency, partnership, or joint venture between JDH and
+Licensee.  This License Agreement does not grant permission to use JDH
+trademarks or trade name in a trademark sense to endorse or promote
+products or services of Licensee, or any third party.
+
+8. By copying, installing or otherwise using matplotlib,
+Licensee agrees to be bound by the terms and conditions of this License
+Agreement.
+
+
+# networkx
+
+NetworkX is distributed with the 3-clause BSD license.
+
+::
+
+   Copyright (C) 2004-2022, NetworkX Developers
+   Aric Hagberg <[email protected]>
+   Dan Schult <[email protected]>
+   Pieter Swart <[email protected]>
+   All rights reserved.
+
+   Redistribution and use in source and binary forms, with or without
+   modification, are permitted provided that the following conditions are
+   met:
+
+     * Redistributions of source code must retain the above copyright
+       notice, this list of conditions and the following disclaimer.
+
+     * Redistributions in binary form must reproduce the above
+       copyright notice, this list of conditions and the following
+       disclaimer in the documentation and/or other materials provided
+       with the distribution.
+
+     * Neither the name of the NetworkX Developers nor the names of its
+       contributors may be used to endorse or promote products derived
+       from this software without specific prior written permission.
+
+   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+# numpy
+
+Copyright (c) 2005-2022, NumPy Developers.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+    * Redistributions of source code must retain the above copyright
+       notice, this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above
+       copyright notice, this list of conditions and the following
+       disclaimer in the documentation and/or other materials provided
+       with the distribution.
+
+    * Neither the name of the NumPy Developers nor the names of any
+       contributors may be used to endorse or promote products derived
+       from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+# opencv-python
+
+MIT License
+
+Copyright (c) Olli-Pekka Heinisuo
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+
+
+# pyparsing
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
+CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
+TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
+SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+
+# python-dateutil
+
+Copyright 2017- Paul Ganssle <[email protected]>
+Copyright 2017- dateutil contributors (see AUTHORS file)
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.
+
+The above license applies to all contributions after 2017-12-01, as well as
+all contributions that have been re-licensed (see AUTHORS file for the list of
+contributors who have re-licensed their code).
+--------------------------------------------------------------------------------
+dateutil - Extensions to the standard Python datetime module.
+
+Copyright (c) 2003-2011 - Gustavo Niemeyer <[email protected]>
+Copyright (c) 2012-2014 - Tomi Pieviläinen <[email protected]>
+Copyright (c) 2014-2016 - Yaron de Leeuw <[email protected]>
+Copyright (c) 2015-     - Paul Ganssle <[email protected]>
+Copyright (c) 2015-     - dateutil contributors (see AUTHORS file)
+
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+    * Redistributions in binary form must reproduce the above copyright notice,
+      this list of conditions and the following disclaimer in the documentation
+      and/or other materials provided with the distribution.
+    * Neither the name of the copyright holder nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
+CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
+PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+The above BSD License Applies to all code, even that also covered by Apache 2.0.
+
+
+# scikit-image
+
+Copyright (C) 2019, the scikit-image team
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+ 1. Redistributions of source code must retain the above copyright
+    notice, this list of conditions and the following disclaimer.
+ 2. Redistributions in binary form must reproduce the above copyright
+    notice, this list of conditions and the following disclaimer in
+    the documentation and/or other materials provided with the
+    distribution.
+ 3. Neither the name of skimage nor the names of its contributors may be
+    used to endorse or promote products derived from this software without
+    specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
+IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT,
+INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
+STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
+IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+
+skimage/_shared/version_requirements.py:_check_version
+
+    Copyright (c) 2013 The IPython Development Team
+    All rights reserved.
+
+    Redistribution and use in source and binary forms, with or without
+    modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice, this
+      list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright notice,
+      this list of conditions and the following disclaimer in the documentation
+      and/or other materials provided with the distribution.
+
+    * Neither the name of the copyright holder nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+    THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+    AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+    IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+    DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+    FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+    DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+    SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+    CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+    OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+    OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+skimage/_shared/version_requirements.py:is_installed:
+
+    Original Copyright (C) 2009-2011 Pierre Raybaut
+
+    Permission is hereby granted, free of charge, to any person obtaining
+    a copy of this software and associated documentation files (the
+    "Software"), to deal in the Software without restriction, including
+    without limitation the rights to use, copy, modify, merge, publish,
+    distribute, sublicense, and/or sell copies of the Software, and to
+    permit persons to whom the Software is furnished to do so, subject to
+    the following conditions:
+
+    The above copyright notice and this permission notice shall be
+    included in all copies or substantial portions of the Software.
+
+    THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+    EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+    MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+    NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+    LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+    OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+    WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+
+# scipy
+
+Copyright (c) 2001-2002 Enthought, Inc. 2003-2022, SciPy Developers.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions
+are met:
+
+1. Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+
+2. Redistributions in binary form must reproduce the above
+   copyright notice, this list of conditions and the following
+   disclaimer in the documentation and/or other materials provided
+   with the distribution.
+
+3. Neither the name of the copyright holder nor the names of its
+   contributors may be used to endorse or promote products derived
+   from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+# six
+
+Copyright (c) 2010-2020 Benjamin Peterson
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of
+this software and associated documentation files (the "Software"), to deal in
+the Software without restriction, including without limitation the rights to
+use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of
+the Software, and to permit persons to whom the Software is furnished to do so,
+subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
+FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR
+COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER
+IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+
+# tiffile
+
+BSD 3-Clause License
+
+Copyright (c) 2008-2022, Christoph Gohlke
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+1. Redistributions of source code must retain the above copyright notice,
+   this list of conditions and the following disclaimer.
+
+2. Redistributions in binary form must reproduce the above copyright notice,
+   this list of conditions and the following disclaimer in the documentation
+   and/or other materials provided with the distribution.
+
+3. Neither the name of the copyright holder nor the names of its
+   contributors may be used to endorse or promote products derived from
+   this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.

src/deskew_HT/README.md

@@ -0,0 +1,75 @@
+# NDLOCR用資料画像の傾き補正モジュール
+
+画像の傾きを補正するモジュールのリポジトリです。
+
+本プログラムは、国立国会図書館が株式会社モルフォAIソリューションズに委託して作成したものです。
+
+本プログラム内の[alyn3](alyn3)は以下のリポジトリのコードを参照し、python3化・高速化等を行い作成しました。
+
+[kakul/Alyn/alyn](https://github.com/kakul/Alyn)
+
+本プログラムの新規開発部分は、国立国会図書館がCC BY 4.0ライセンスで公開するものです。詳細については
+[LICENSE](./LICENSE
+)をご覧ください。
+
+# 概要
+
+入力画像の直線要素を検出することで画像の傾きを推定し、傾きを補正した画像を出力する。
+
+出力画像は元画像の領域が欠損しないように拡大して保存する仕様となっており、
+この際に元画像領域外の部分は指定した濃さのグレー(デフォルトは白)で塗りつぶされる。
+
+推定した傾きの数値情報は、オプション（後述）を指定することでテキストファイルとして出力できる。
+
+
+# 使い方
+
+指定パスの入力画像または指定ディレクトリ内の画像の傾きを推定し補正する。
+
+※補正角度は±45度以内に限る。
+
+```
+python3 run_deskew.py INPUT [-o OUTPUT] [-s SKEW_MAX] [-a ANGLE_ACC] [-m METHOD]
+```
+
+positional arguments:
+```
+  input                 入力画像のパス、または入力画像を格納したディレクトリのパス
+```
+
+optional arguments:
+```
+  -h, --help            ヘルプメッセージを表示して終了
+  -o OUT, --out OUT     出力ファイルのパス(INPUTが画像ファイルの時、default: out.jpg)または
+                        出力ディレクトリのパス(INPUTがディレクトリの時、default: out)
+  -l LOG, --log LOG     推定した傾きを保存するテキストファイルのパス。指定なしの場合出力されない
+                        処理画像一枚ごとに次の形式で指定ファイルの最終行に追加する。
+                        output format:
+                        Image_file_path <tab> Estimated_skew_angle[deg]
+  -s SKEW_MAX, --skew_max SKEW_MAX
+                        推定する傾きの最大角度[deg] default: 4.0[deg]
+                        0より大きい45以下の値を指定する。大きくするほど処理時間は増加
+  -a ANGLE_ACC, --angle_acc ANGLE_ACC
+                        傾きの探索を何度単位で行うか。default: 0.5[deg]
+                        0より大きいSKEW_MAX以下の値を指定する。小さくするほど処理時間は増加。
+  -rw ROI_WIDTH, --roi_width ROI_WIDTH
+                        直線検出の対象とする関心領域の画像全体に対する水平方向の割合
+                        0.0より大きい1.0以下の数 default: 1.0(水平方向全体)
+  -rh ROI_HEIGHT, --roi_height ROI_HEIGHT
+                        直線検出の対象とする関心領域の画像全体に対する鉛直方向の割合
+                        0.0より大きい1.0以下の数 default: 1.0(鉛直方向全体)
+  -m METHOD, --method METHOD
+                        画像回転時の補完手法。以下の整数値で指定する。
+                        0: Nearest-neighbor  1: Bi-linear(default)
+                        2: Bi-quadratic      3: Bi-cubic
+                        4: Bi-quartic        5: Bi-quintic
+  -g GRAY, --gray GRAY  出力画像において、元画像領域の外側を補完するグレーの濃さ
+                        0(黒) 以上 1.0(白)以下で指定する。default: 1.0(白)
+  -q QUALITY, --quality QUALITY
+                        Jpeg画像出力時の画質。
+                        1が最低画質で最小ファイルサイズ、100が最高画質で最大ファイルサイズ。
+                        [1, 100], default: 100
+  --short SHORT         出力画像の短辺の長さ。アスペクト比は維持したままリサイズする。
+                        指定しなかった場合オリジナルサイズで出力される。
+  -v, --version         プログラムのバージョンを表示して終了
+```

src/deskew_HT/alyn3/deskew.py

@@ -0,0 +1,163 @@
+""" Deskews file after getting skew angle """
+"""
+This code is based on the following file:
+https://github.com/kakul/Alyn/blob/master/alyn/deskew.py
+"""
+import optparse
+import numpy as np
+import os
+
+from alyn3.skew_detect import SkewDetect
+import cv2
+
+
+class Deskew:
+
+    def __init__(self, input_file, output_file, r_angle=0,
+                 skew_max=4.0, acc_deg=0.1, method=1,
+                 roi_w=1.0, roi_h=1.0,
+                 gray=1.0, quality=100, short=None):
+        self.input_file = input_file
+        self.output_file = output_file
+        self.r_angle = r_angle
+        self.method = method
+        self.gray = gray
+        self.quality = quality
+        self.short = short
+        self.skew_obj = SkewDetect(self.input_file,
+                                   skew_max=skew_max, acc_deg=acc_deg,
+                                   roi_w=roi_w, roi_h=roi_h)
+
+    def deskew(self):
+        print('input: '+self.input_file)
+
+        res = self.skew_obj.process_single_file()
+        angle = res['Estimated Angle']
+        rot_angle = angle + self.r_angle
+
+        img = cv2.imread(self.input_file, cv2.IMREAD_COLOR)
+        g = self.gray * 255
+        rotated = self.rotate_expand(img, rot_angle, g)
+
+        if self.short:
+            h = rotated.shape[0]
+            w = rotated.shape[1]
+            print('origin w,h: {}, {}'.format(w, h))
+            if w < h:
+                h = int(h*self.short/w+0.5)
+                w = self.short
+            else:
+                w = int(w*self.short/h+0.5)
+                h = self.short
+            print('resized w,h: {}, {}'.format(w, h))
+            rotated = cv2.resize(rotated, (w, h))
+
+        if self.output_file:
+            self.save_image(rotated)
+
+        return res
+
+    def deskew_on_memory(self, input_data):
+        res = self.skew_obj.determine_skew_on_memory(input_data)
+        angle = res['Estimated Angle']
+        rot_angle = angle + self.r_angle
+
+        img = input_data
+        g = self.gray * 255
+        rotated = self.rotate_expand(img, rot_angle, g)
+
+        if self.short:
+            h = rotated.shape[0]
+            w = rotated.shape[1]
+            print('origin w,h: {}, {}'.format(w, h))
+            if w < h:
+                h = int(h*self.short/w+0.5)
+                w = self.short
+            else:
+                w = int(w*self.short/h+0.5)
+                h = self.short
+            print('resized w,h: {}, {}'.format(w, h))
+            rotated = cv2.resize(rotated, (w, h))
+
+        return rotated
+
+    def save_image(self, img):
+        path = self.skew_obj.check_path(self.output_file)
+        if os.path.splitext(path)[1] in ['.jpg', '.JPG', '.jpeg', '.JPEG']:
+            cv2.imwrite(path, img, [cv2.IMWRITE_JPEG_QUALITY, 100])
+        else:
+            cv2.imwrite(path, img)
+
+    def rotate_expand(self, img, angle=0, g=255):
+        h = img.shape[0]
+        w = img.shape[1]
+        angle_rad = angle/180.0*np.pi
+        w_rot = int(np.round(h*np.absolute(np.sin(angle_rad)) +
+                    w*np.absolute(np.cos(angle_rad))))
+        h_rot = int(np.round(h*np.absolute(np.cos(angle_rad)) +
+                    w*np.absolute(np.sin(angle_rad))))
+        size_rot = (w_rot, h_rot)
+        mat = cv2.getRotationMatrix2D((w/2, h/2), angle, 1.0)
+        mat[0][2] = mat[0][2] - w/2 + w_rot/2
+        mat[1][2] = mat[1][2] - h/2 + h_rot/2
+        rotated = cv2.warpAffine(img, mat, size_rot, borderValue=(g, g, g))
+
+        return rotated
+
+    def run(self):
+        if self.input_file:
+            return self.deskew()
+
+
+def optparse_args():
+    parser = optparse.OptionParser()
+
+    parser.add_option(
+        '-i',
+        '--input',
+        default=None,
+        dest='input_file',
+        help='Input file name')
+    parser.add_option(
+        '-o', '--output',
+        default=None,
+        dest='output_file',
+        help='Output file name')
+    parser.add_option(
+        '-r', '--rotate',
+        default=0,
+        dest='r_angle',
+        help='Rotate the image to desired axis',
+        type=int)
+    parser.add_option(
+        '-g', '--gray',
+        default=1.0,
+        dest='gray',
+        help='Gray level outside the input image boundaries.\n'
+             'between 0.0(black) and 1.0(white)\n'
+             '[0.0, 1.0], default: 1.0',
+        type=float)
+    parser.add_option(
+        '-q', '--quality',
+        default=100,
+        dest='quality',
+        help='output jpeg image quality. i\n'
+             '1 is worst quality and smallest file size,\n'
+             'and 100 is best quality and largest file size.\n'
+             '[1, 100], default: 100',
+        type=int)
+
+    return parser.parse_args()
+
+
+if __name__ == '__main__':
+    options, args = optparse_args()
+    deskew_obj = Deskew(
+        options.input_file,
+        options.display_image,
+        options.output_file,
+        options.r_angle,
+        options.gray,
+        options.quality)
+
+    deskew_obj.run()

src/deskew_HT/alyn3/skew_detect.py

@@ -0,0 +1,388 @@
+""" Calculates skew angle """
+"""
+This code is based on the following file:
+https://github.com/kakul/Alyn/blob/master/alyn/skew_detect.py
+"""
+import os
+import optparse
+
+import numpy as np
+# import matplotlib.pyplot as plt
+from skimage import io
+from skimage.feature import canny
+from skimage.transform import hough_line, hough_line_peaks
+import cv2
+
+
+class SkewDetect:
+
+    piby4 = np.pi / 4
+
+    def __init__(
+        self,
+        input_file=None,
+        output_file=None,
+        sigma=0.50,
+        display_output=None,
+        num_peaks=20,
+        skew_max=4.0,
+        acc_deg=0.5,
+        roi_w=1.0,
+        roi_h=1.0,
+    ):
+
+        self.sigma = sigma
+        self.input_file = input_file
+        self.output_file = output_file
+        self.display_output = display_output
+        self.num_peaks = num_peaks
+        self.skew_max = skew_max
+        self.acc_deg = acc_deg
+        self.roi_w = roi_w
+        self.roi_h = roi_h
+
+    def write_to_file(self, wfile, data):
+
+        for d in data:
+            wfile.write(d + ': ' + str(data[d]) + '\n')
+        wfile.write('\n')
+
+    def get_max_freq_elem(self, arr):
+
+        max_arr = []
+        freqs = {}
+        for i in arr:
+            if i in freqs:
+                freqs[i] += 1
+            else:
+                freqs[i] = 1
+
+        sorted_keys = sorted(freqs, key=freqs.get, reverse=True)
+        max_freq = freqs[sorted_keys[0]]
+
+        for k in sorted_keys:
+            if freqs[k] == max_freq:
+                max_arr.append(k)
+
+        return max_arr
+
+    def compare_sum(self, value):
+        if value >= 44 and value <= 46:
+            return True
+        else:
+            return False
+
+    def display(self, data):
+
+        for i in data:
+            print(str(i) + ": " + str(data[i]))
+
+    def calculate_deviation(self, angle):
+
+        angle_in_degrees = np.abs(angle)
+        deviation = np.abs(SkewDetect.piby4 - angle_in_degrees)
+
+        return deviation
+
+    def run(self):
+
+        if self.display_output:
+            if self.display_output.lower() == 'yes':
+                self.display_output = True
+            else:
+                self.display_output = False
+
+        if self.input_file is None:
+            print("Invalid input, nothing to process.")
+        else:
+            self.process_single_file()
+
+    def check_path(self, path):
+
+        if os.path.isabs(path):
+            full_path = path
+        else:
+            full_path = os.getcwd() + '/' + str(path)
+        return full_path
+
+    def process_single_file(self):
+
+        file_path = self.check_path(self.input_file)
+        res = self.determine_skew(file_path)
+
+        if self.output_file:
+            output_path = self.check_path(self.output_file)
+            wfile = open(output_path, 'w')
+            self.write_to_file(wfile, res)
+            wfile.close()
+
+        return res
+
+    def determine_skew(self, img_file):
+
+        img_ori = io.imread(img_file, as_gray=True)
+        height, width = img_ori.shape
+        img = img_ori[int(height*(0.5-self.roi_h/2.0)):int(height*(0.5+self.roi_h/2.0)),
+                      int(width * (0.5-self.roi_w/2.0)):int(width * (0.5+self.roi_w/2.0))]
+
+        img = cv2.resize(img, (img.shape[1]//4, img.shape[0]//4))
+
+        edges = canny(img, sigma=self.sigma)
+        range_rad = np.arange(-np.pi/2, -np.pi/2+np.deg2rad(self.skew_max),
+                              step=np.deg2rad(self.acc_deg))
+        range_rad = np.concatenate(
+            [range_rad,
+             np.arange(-np.deg2rad(self.skew_max), np.deg2rad(self.skew_max),
+                       step=np.deg2rad(self.acc_deg))],
+            axis=0)
+        range_rad = np.concatenate(
+            [range_rad,
+             np.arange(np.pi/2-np.deg2rad(self.skew_max), np.pi/2,
+                       step=np.deg2rad(self.acc_deg))],
+            axis=0)
+
+        h, a, d = hough_line(edges, theta=range_rad)
+
+        th = 0.2 * h.max()
+        _, ap, _ = hough_line_peaks(
+            h, a, d, threshold=th, num_peaks=self.num_peaks)
+
+        if len(ap) == 0:
+            data = {
+                "Image File": img_file,
+                "Average Deviation from pi/4": 0.0,
+                "Estimated Angle": 0.0,
+                "Angle bins": [[], [], [], []],
+                "Message": "Bad Quality"}
+            return data
+
+        absolute_deviations = [self.calculate_deviation(k) for k in ap]
+        average_deviation = np.mean(np.rad2deg(absolute_deviations))
+        ap_deg = [np.rad2deg(x) for x in ap]
+
+        for i in range(len(ap_deg)):
+            if ap_deg[i] >= 45.0:
+                ap_deg[i] -= 90.0
+            elif ap_deg[i] <= -45.0:
+                ap_deg[i] += 90.0
+
+        bin_0_45 = []
+        bin_45_90 = []
+        bin_0_45n = []
+        bin_45_90n = []
+
+        for ang in ap_deg:
+
+            deviation_sum = (90 - ang + average_deviation)
+            if self.compare_sum(deviation_sum):
+                bin_45_90.append(ang)
+                continue
+
+            deviation_sum = (ang + average_deviation)
+            if self.compare_sum(deviation_sum):
+                bin_0_45.append(ang)
+                continue
+
+            deviation_sum = (-ang + average_deviation)
+            if self.compare_sum(deviation_sum):
+                bin_0_45n.append(ang)
+                continue
+
+            deviation_sum = (90 + ang + average_deviation)
+            if self.compare_sum(deviation_sum):
+                bin_45_90n.append(ang)
+
+        angles = [bin_0_45, bin_45_90, bin_0_45n, bin_45_90n]
+        lmax = 0
+
+        for j in range(len(angles)):
+            tmp_l = len(angles[j])
+            if tmp_l > lmax:
+                lmax = tmp_l
+                maxi = j
+
+        if lmax:
+            ans_arr = self.get_max_freq_elem(angles[maxi])  # 最多頻度の角度array
+            ans_res = np.mean(ans_arr)    # 同数最多が複数あるかもしれないのでavg
+
+        else:  # angls が空のとき
+            ans_arr = self.get_max_freq_elem(ap_deg)
+            ans_res = np.mean(ans_arr)
+
+        data = {
+            "Image File": img_file,
+            "Average Deviation from pi/4": average_deviation,
+            "Estimated Angle": ans_res,
+            "Angle bins": angles,
+            "Message": "Successfully detected lines"}
+
+        if self.display_output:
+            self.display(data)
+
+        return data
+
+    def determine_skew_on_memory(self, img_data):
+
+        img_ori = cv2.cvtColor(img_data, cv2.COLOR_BGR2GRAY)
+        height, width = img_ori.shape
+        img = img_ori[int(height*(0.5-self.roi_h/2.0)):int(height*(0.5+self.roi_h/2.0)),
+                      int(width * (0.5-self.roi_w/2.0)):int(width * (0.5+self.roi_w/2.0))]
+
+        img = cv2.resize(img, (img.shape[1]//4, img.shape[0]//4))
+
+        edges = canny(img, sigma=self.sigma)
+        range_rad = np.arange(-np.pi/2, -np.pi/2+np.deg2rad(self.skew_max),
+                              step=np.deg2rad(self.acc_deg))
+        range_rad = np.concatenate([range_rad,
+                                    np.arange(-np.deg2rad(self.skew_max),
+                                              np.deg2rad(self.skew_max),
+                                              step=np.deg2rad(self.acc_deg))],
+                                   axis=0)
+        range_rad = np.concatenate([range_rad,
+                                    np.arange(np.pi/2-np.deg2rad(self.skew_max),
+                                              np.pi/2,
+                                              step=np.deg2rad(self.acc_deg))],
+                                   axis=0)
+
+        h, a, d = hough_line(edges, theta=range_rad)
+
+        th = 0.2 * h.max()
+        _, ap, _ = hough_line_peaks(
+            h, a, d, threshold=th, num_peaks=self.num_peaks)
+
+        if len(ap) == 0:
+            data = {
+                "Average Deviation from pi/4": 0.0,
+                "Estimated Angle": 0.0,
+                "Angle bins": [[], [], [], []],
+                "Message": "Bad Quality"}
+            return data
+
+        absolute_deviations = [self.calculate_deviation(k) for k in ap]
+        average_deviation = np.mean(np.rad2deg(absolute_deviations))
+        ap_deg = [np.rad2deg(x) for x in ap]
+
+        for i in range(len(ap_deg)):
+            if ap_deg[i] >= 45.0:
+                ap_deg[i] -= 90.0
+            elif ap_deg[i] <= -45.0:
+                ap_deg[i] += 90.0
+
+        bin_0_45 = []
+        bin_45_90 = []
+        bin_0_45n = []
+        bin_45_90n = []
+
+        for ang in ap_deg:
+
+            deviation_sum = (90 - ang + average_deviation)
+            if self.compare_sum(deviation_sum):
+                bin_45_90.append(ang)
+                continue
+
+            deviation_sum = (ang + average_deviation)
+            if self.compare_sum(deviation_sum):
+                bin_0_45.append(ang)
+                continue
+
+            deviation_sum = (-ang + average_deviation)
+            if self.compare_sum(deviation_sum):
+                bin_0_45n.append(ang)
+                continue
+
+            deviation_sum = (90 + ang + average_deviation)
+            if self.compare_sum(deviation_sum):
+                bin_45_90n.append(ang)
+
+        angles = [bin_0_45, bin_45_90, bin_0_45n, bin_45_90n]
+        lmax = 0
+
+        for j in range(len(angles)):
+            tmp_l = len(angles[j])
+            if tmp_l > lmax:
+                lmax = tmp_l
+                maxi = j
+
+        if lmax:
+            ans_arr = self.get_max_freq_elem(angles[maxi])  # 最多頻度の角度array
+            ans_res = np.mean(ans_arr)    # 同数最多が複数あるかもしれないのでavg
+
+        else:  # angls が空のとき
+            ans_arr = self.get_max_freq_elem(ap_deg)
+            ans_res = np.mean(ans_arr)
+
+        data = {
+            "Average Deviation from pi/4": average_deviation,
+            "Estimated Angle": ans_res,
+            "Angle bins": angles,
+            "Message": "Successfully detected lines"}
+
+        return data
+
+    def determine_line(self, img_file):
+
+        img_ori = io.imread(img_file, as_gray=True)
+        height, width = img_ori.shape
+        img = img_ori[int(height*(0.5-self.roi_h/2.0)):int(height*(0.5+self.roi_h/2.0)),
+                      int(width * (0.5-self.roi_w/2.0)):int(width * (0.5+self.roi_w/2.0))]
+        edges = canny(img, sigma=self.sigma)
+        range_rad = np.arange(-np.pi/2, -np.pi/2+np.deg2rad(self.skew_max),
+                              step=np.deg2rad(self.acc_deg))
+        range_rad = np.concatenate([range_rad,
+                                    np.arange(-np.deg2rad(self.skew_max),
+                                              np.deg2rad(self.skew_max),
+                                              step=np.deg2rad(self.acc_deg))],
+                                   axis=0)
+        range_rad = np.concatenate([range_rad,
+                                    np.arange(np.pi/2-np.deg2rad(self.skew_max), np.pi/2,
+                                              step=np.deg2rad(self.acc_deg))],
+                                   axis=0)
+
+        h, a, d = hough_line(edges, theta=range_rad)
+
+        th = 0.2 * h.max()
+        ac, ap, d = hough_line_peaks(
+            h, a, d, threshold=th, num_peaks=self.num_peaks)
+
+        return ac, ap, d
+
+
+if __name__ == '__main__':
+
+    parser = optparse.OptionParser()
+
+    parser.add_option(
+        '-d', '--display',
+        default=None,
+        dest='display_output',
+        help='Display logs')
+    parser.add_option(
+        '-i', '--input',
+        default=None,
+        dest='input_file',
+        help='Input file name')
+    parser.add_option(
+        '-o', '--output',
+        default=None,
+        dest='output_file',
+        help='Output file name')
+    parser.add_option(
+        '-p', '--plot',
+        default=None,
+        dest='plot_hough',
+        help='Plot the Hough Transform')
+    parser.add_option(
+        '-s', '--sigma',
+        default=3.0,
+        dest='sigma',
+        help='Sigma for Canny Edge Detection',
+        type=float)
+    options, args = parser.parse_args()
+    skew_obj = SkewDetect(
+        options.input_file,
+        options.output_file,
+        options.sigma,
+        options.display_output,
+        options.num_peaks,
+        options.plot_hough)
+    skew_obj.run()

src/deskew_HT/run_deskew.py

@@ -0,0 +1,272 @@
+# !/usr/bin/env python3
+
+# Copyright (c) 2022, National Diet Library, Japan
+#
+# This software is released under the CC BY 4.0.
+# https://creativecommons.org/licenses/by/4.0/
+
+import os
+import cv2
+import argparse
+import numpy as np
+from alyn3.deskew import Deskew
+from alyn3.skew_detect import SkewDetect
+
+os.environ["OPENCV_IO_ENABLE_JASPER"] = "true"
+
+
+def deskew_image(input, output, r_angle=0,
+                 skew_max=4.0, acc_deg=0.5, roi_w=1.0, roi_h=1.0,
+                 method=1, gray=1.0, quality=100, short=None,
+                 log=None):
+
+    image_name = os.path.basename(input)
+    print('process: '+image_name)
+    d = Deskew(input, output,
+               r_angle=r_angle,
+               skew_max=skew_max,
+               acc_deg=acc_deg,
+               method=method,
+               gray=gray,
+               quality=quality,
+               short=short,
+               roi_w=roi_w,
+               roi_h=roi_h)
+    res = d.run()
+
+    if log:
+        with open(log, mode='a') as f:
+            line = '{}\t{:.6f}\n'.format(
+                res['Image File'], (-res['Estimated Angle']))
+            f.write(line)
+
+
+def deskew_dir(input_dir_path, output_dir_path, r_angle=0,
+               skew_max=4.0, acc_deg=0.5, roi_w=1.0, roi_h=1.0,
+               method=1, gray=1.0, quality=100, short=None,
+               log=None):
+    image_list = os.listdir(input_dir_path)
+
+    for image_name in image_list:
+        input_path = os.path.join(input_dir_path, image_name)
+        if(os.path.isdir(input_path)):
+            continue
+        print('process: '+str(image_name))
+        output_path = os.path.join(output_dir_path, image_name)
+        d = Deskew(input_path, output_path,
+                   r_angle=r_angle,
+                   skew_max=skew_max,
+                   acc_deg=acc_deg,
+                   method=method,
+                   gray=gray,
+                   quality=quality,
+                   short=short,
+                   roi_w=roi_w,
+                   roi_h=roi_h)
+        res = d.run()
+
+        if log:
+            with open(log, mode='a') as f:
+                line = '{}\t{:.6f}\n'.format(
+                    res['Image File'], (-res['Estimated Angle']))
+                f.write(line)
+
+
+def add_detected_lines(input_path, output_path,
+                       skew_max=4.0, acc_deg=0.5,
+                       roi_w=1.0, roi_h=1.0,
+                       bgr=[0, 0, 255]):
+    line_len = 4000
+
+    print("Add the detected lines to "+os.path.basename(input_path))
+    sd = SkewDetect(input_path, skew_max=skew_max, acc_deg=acc_deg,
+                    roi_w=roi_w, roi_h=roi_h)
+    acc, ang_rad, distance = sd.determine_line(input_path)
+    img = cv2.imread(input_path)
+    if len(acc) == 0:
+        print('Image file:{} has no lines detected'.format(input_path))
+    else:
+        max_val = max(acc)
+        for val, theta, rho in zip(acc[::-1], ang_rad[::-1], distance[::-1]):
+            a = np.cos(theta)
+            b = np.sin(theta)
+            x0 = a * rho + int(img.shape[1] * (0.5-roi_w/2.0))
+            y0 = b * rho + int(img.shape[0] * (0.5-roi_h/2.0))
+            x1 = int(x0 + line_len*(-b))
+            y1 = int(y0 + line_len*(a))
+            x2 = int(x0 - line_len*(-b))
+            y2 = int(y0 - line_len*(a))
+            tmp_bgr = bgr.copy()
+            tmp_bgr[0] = 255.0 * (1.0 - val / max_val)
+            tmp_bgr[1] = tmp_bgr[0]
+            # print(tmp_bgr)
+            cv2.line(img, (x1, y1), (x2, y2), tmp_bgr, 2)
+
+    cv2.imwrite(output_path, img)
+
+
+def add_detected_lines_dir(input_dir_path, output_dir_path,
+                           skew_max=4.0, acc_deg=0.1,
+                           roi_w=1.0, roi_h=1.0,
+                           bgr=[0, 0, 255]):
+    # Hough変換で検知したLineを元画像に書き加える
+    # Add the lines detected by Hough Transform to the input images
+    image_list = os.listdir(input_dir_path)
+
+    for image_name in image_list:
+        input_path = os.path.join(input_dir_path, image_name)
+        if(os.path.isdir(input_path)):
+            continue
+        output_path = os.path.join(output_dir_path, image_name)
+        add_detected_lines(input_path, output_path,
+                           skew_max=skew_max, acc_deg=acc_deg,
+                           roi_w=roi_w, roi_h=roi_h,
+                           bgr=[0, 0, 255])
+
+
+def parse_args():
+    usage = 'python3 {} INPUT [-o OUTPUT] [-s SKEW_MAX] [-a ANGLE_ACC] [-m METHOD]'.format(
+        __file__)
+    argparser = argparse.ArgumentParser(
+        usage=usage,
+        description='Deskew image(when INPUT is an image) or images in INPUT(when INPUT is a directory).',
+        formatter_class=argparse.RawTextHelpFormatter)
+    argparser.add_argument(
+        'input',
+        help='input image file or directory path',
+        type=str)
+    argparser.add_argument(
+        '-o',
+        '--out',
+        default='out.jpg',
+        help='output file or directory path',
+        type=str)
+    argparser.add_argument(
+        '-l',
+        '--log',
+        default=None,
+        help='estimated skew log file path\n'
+             'output format:\n'
+             'Image_file_path <tab> Estimated_skew_angle[deg]')
+    argparser.add_argument(
+        '-s',
+        '--skew_max',
+        default=4.0,
+        help='maximum expected skew angle[deg], default: 4.0',
+        type=float)
+    argparser.add_argument(
+        '-a',
+        '--angle_acc',
+        default=0.5,
+        help='estimated skew angle accuracy[deg], default: 0.5',
+        type=float)
+    argparser.add_argument(
+        '-rw',
+        '--roi_width',
+        default=1.0,
+        help='horizontal cropping ratio of the region of interest \n'
+             'to the whole image. (0.0, 1.0] default: 1.0(whole image)',
+        type=float)
+    argparser.add_argument(
+        '-rh',
+        '--roi_height',
+        default=1.0,
+        help='vertical cropping ratio of the region of interest \n'
+             'to the whole image. (0.0, 1.0] default: 1.0(whole image)',
+        type=float)
+    argparser.add_argument(
+        '-m',
+        '--method',
+        default=1,
+        help='interpolation method.\n'
+             '0: Nearest-neighbor  1: Bi-linear(default)\n'
+             '2: Bi-quadratic      3: Bi-cubic\n'
+             '4: Bi-quartic        5: Bi-quintic\n',
+        type=int)
+    argparser.add_argument(
+        '-g',
+        '--gray',
+        default=1.0,
+        dest='gray',
+        help='gray value outside the input image boundaries.\n'
+             '[0.0(black), 1.0(white)], default: 1.0',
+        type=float)
+    argparser.add_argument(
+        '-q', '--quality',
+        default=100,
+        dest='quality',
+        help='output jpeg image quality.\n'
+             '1 is worst quality and smallest file size,\n'
+             'and 100 is best quality and largest file size.\n'
+             '[1, 100], default: 100',
+        type=int)
+    argparser.add_argument(
+        '--short',
+        default=None,
+        dest='short',
+        help='the length of the short side of the output image.',
+        type=int)
+    argparser.add_argument(
+        '-v',
+        '--version',
+        version='deskew version 1.0.0',
+        action='version')
+    argparser.add_argument(
+        '--debug',
+        action='store_true')
+
+    return argparser.parse_args()
+
+
+if __name__ == '__main__':
+
+    args = parse_args()
+
+    input = args.input
+    output = args.out
+    print('input directory/image: '+input)
+
+    if(os.path.isdir(input)):  # directory
+        if output[-4:] == '.jpg':
+            output = output[:-4]  # 'out'
+        print('output: '+output)
+        os.makedirs(output, exist_ok=True)
+        deskew_dir(input, output,
+                   r_angle=0,
+                   skew_max=args.skew_max,
+                   acc_deg=args.angle_acc,
+                   roi_w=args.roi_width,
+                   roi_h=args.roi_height,
+                   method=args.method,
+                   gray=args.gray,
+                   quality=args.quality,
+                   short=args.short,
+                   log=args.log)
+        if args.debug:
+            print('[Debug] Dump input images with detected lines')
+            os.makedirs(output+'_withL', exist_ok=True)
+            add_detected_lines_dir(input, output+'_withL',
+                                   roi_w=args.roi_width,
+                                   roi_h=args.roi_height,
+                                   skew_max=args.skew_max,
+                                   acc_deg=args.angle_acc)
+    else:  # single image
+        print('output: '+output)
+        deskew_image(input, output,
+                     r_angle=0,
+                     skew_max=args.skew_max,
+                     acc_deg=args.angle_acc,
+                     roi_w=args.roi_width,
+                     roi_h=args.roi_height,
+                     method=args.method,
+                     gray=args.gray,
+                     quality=args.quality,
+                     short=args.short,
+                     log=args.log)
+        if args.debug:
+            print('[Debug] Dump input image with detected lines')
+            add_detected_lines(input, output+'_withL.jpg',
+                               roi_w=args.roi_width,
+                               roi_h=args.roi_height,
+                               skew_max=args.skew_max,
+                               acc_deg=args.angle_acc)

src/deskew_HT/setup.cfg

@@ -0,0 +1,3 @@
+[flake8]
+extend-ignore = E402, E501, E303
+exclude = 

src/ndl_layout/.gitmodules

@@ -0,0 +1,3 @@
+[submodule "mmdetection"]
+	path = mmdetection
+	url = https://github.com/ndl-lab/mmdetection.git

src/ndl_layout/LICENSE

@@ -0,0 +1,395 @@
+Copyright (c) 2022, National Diet Library, Japan.
+
+Attribution 4.0 International
+
+=======================================================================
+
+Creative Commons Corporation ("Creative Commons") is not a law firm and
+does not provide legal services or legal advice. Distribution of
+Creative Commons public licenses does not create a lawyer-client or
+other relationship. Creative Commons makes its licenses and related
+information available on an "as-is" basis. Creative Commons gives no
+warranties regarding its licenses, any material licensed under their
+terms and conditions, or any related information. Creative Commons
+disclaims all liability for damages resulting from their use to the
+fullest extent possible.
+
+Using Creative Commons Public Licenses
+
+Creative Commons public licenses provide a standard set of terms and
+conditions that creators and other rights holders may use to share
+original works of authorship and other material subject to copyright
+and certain other rights specified in the public license below. The
+following considerations are for informational purposes only, are not
+exhaustive, and do not form part of our licenses.
+
+     Considerations for licensors: Our public licenses are
+     intended for use by those authorized to give the public
+     permission to use material in ways otherwise restricted by
+     copyright and certain other rights. Our licenses are
+     irrevocable. Licensors should read and understand the terms
+     and conditions of the license they choose before applying it.
+     Licensors should also secure all rights necessary before
+     applying our licenses so that the public can reuse the
+     material as expected. Licensors should clearly mark any
+     material not subject to the license. This includes other CC-
+     licensed material, or material used under an exception or
+     limitation to copyright. More considerations for licensors:
+	wiki.creativecommons.org/Considerations_for_licensors
+
+     Considerations for the public: By using one of our public
+     licenses, a licensor grants the public permission to use the
+     licensed material under specified terms and conditions. If
+     the licensor's permission is not necessary for any reason--for
+     example, because of any applicable exception or limitation to
+     copyright--then that use is not regulated by the license. Our
+     licenses grant only permissions under copyright and certain
+     other rights that a licensor has authority to grant. Use of
+     the licensed material may still be restricted for other
+     reasons, including because others have copyright or other
+     rights in the material. A licensor may make special requests,
+     such as asking that all changes be marked or described.
+     Although not required by our licenses, you are encouraged to
+     respect those requests where reasonable. More_considerations
+     for the public: 
+	wiki.creativecommons.org/Considerations_for_licensees
+
+=======================================================================
+
+Creative Commons Attribution 4.0 International Public License
+
+By exercising the Licensed Rights (defined below), You accept and agree
+to be bound by the terms and conditions of this Creative Commons
+Attribution 4.0 International Public License ("Public License"). To the
+extent this Public License may be interpreted as a contract, You are
+granted the Licensed Rights in consideration of Your acceptance of
+these terms and conditions, and the Licensor grants You such rights in
+consideration of benefits the Licensor receives from making the
+Licensed Material available under these terms and conditions.
+
+
+Section 1 -- Definitions.
+
+  a. Adapted Material means material subject to Copyright and Similar
+     Rights that is derived from or based upon the Licensed Material
+     and in which the Licensed Material is translated, altered,
+     arranged, transformed, or otherwise modified in a manner requiring
+     permission under the Copyright and Similar Rights held by the
+     Licensor. For purposes of this Public License, where the Licensed
+     Material is a musical work, performance, or sound recording,
+     Adapted Material is always produced where the Licensed Material is
+     synched in timed relation with a moving image.
+
+  b. Adapter's License means the license You apply to Your Copyright
+     and Similar Rights in Your contributions to Adapted Material in
+     accordance with the terms and conditions of this Public License.
+
+  c. Copyright and Similar Rights means copyright and/or similar rights
+     closely related to copyright including, without limitation,
+     performance, broadcast, sound recording, and Sui Generis Database
+     Rights, without regard to how the rights are labeled or
+     categorized. For purposes of this Public License, the rights
+     specified in Section 2(b)(1)-(2) are not Copyright and Similar
+     Rights.
+
+  d. Effective Technological Measures means those measures that, in the
+     absence of proper authority, may not be circumvented under laws
+     fulfilling obligations under Article 11 of the WIPO Copyright
+     Treaty adopted on December 20, 1996, and/or similar international
+     agreements.
+
+  e. Exceptions and Limitations means fair use, fair dealing, and/or
+     any other exception or limitation to Copyright and Similar Rights
+     that applies to Your use of the Licensed Material.
+
+  f. Licensed Material means the artistic or literary work, database,
+     or other material to which the Licensor applied this Public
+     License.
+
+  g. Licensed Rights means the rights granted to You subject to the
+     terms and conditions of this Public License, which are limited to
+     all Copyright and Similar Rights that apply to Your use of the
+     Licensed Material and that the Licensor has authority to license.
+
+  h. Licensor means the individual(s) or entity(ies) granting rights
+     under this Public License.
+
+  i. Share means to provide material to the public by any means or
+     process that requires permission under the Licensed Rights, such
+     as reproduction, public display, public performance, distribution,
+     dissemination, communication, or importation, and to make material
+     available to the public including in ways that members of the
+     public may access the material from a place and at a time
+     individually chosen by them.
+
+  j. Sui Generis Database Rights means rights other than copyright
+     resulting from Directive 96/9/EC of the European Parliament and of
+     the Council of 11 March 1996 on the legal protection of databases,
+     as amended and/or succeeded, as well as other essentially
+     equivalent rights anywhere in the world.
+
+  k. You means the individual or entity exercising the Licensed Rights
+     under this Public License. Your has a corresponding meaning.
+
+
+Section 2 -- Scope.
+
+  a. License grant.
+
+       1. Subject to the terms and conditions of this Public License,
+          the Licensor hereby grants You a worldwide, royalty-free,
+          non-sublicensable, non-exclusive, irrevocable license to
+          exercise the Licensed Rights in the Licensed Material to:
+
+            a. reproduce and Share the Licensed Material, in whole or
+               in part; and
+
+            b. produce, reproduce, and Share Adapted Material.
+
+       2. Exceptions and Limitations. For the avoidance of doubt, where
+          Exceptions and Limitations apply to Your use, this Public
+          License does not apply, and You do not need to comply with
+          its terms and conditions.
+
+       3. Term. The term of this Public License is specified in Section
+          6(a).
+
+       4. Media and formats; technical modifications allowed. The
+          Licensor authorizes You to exercise the Licensed Rights in
+          all media and formats whether now known or hereafter created,
+          and to make technical modifications necessary to do so. The
+          Licensor waives and/or agrees not to assert any right or
+          authority to forbid You from making technical modifications
+          necessary to exercise the Licensed Rights, including
+          technical modifications necessary to circumvent Effective
+          Technological Measures. For purposes of this Public License,
+          simply making modifications authorized by this Section 2(a)
+          (4) never produces Adapted Material.
+
+       5. Downstream recipients.
+
+            a. Offer from the Licensor -- Licensed Material. Every
+               recipient of the Licensed Material automatically
+               receives an offer from the Licensor to exercise the
+               Licensed Rights under the terms and conditions of this
+               Public License.
+
+            b. No downstream restrictions. You may not offer or impose
+               any additional or different terms or conditions on, or
+               apply any Effective Technological Measures to, the
+               Licensed Material if doing so restricts exercise of the
+               Licensed Rights by any recipient of the Licensed
+               Material.
+
+       6. No endorsement. Nothing in this Public License constitutes or
+          may be construed as permission to assert or imply that You
+          are, or that Your use of the Licensed Material is, connected
+          with, or sponsored, endorsed, or granted official status by,
+          the Licensor or others designated to receive attribution as
+          provided in Section 3(a)(1)(A)(i).
+
+  b. Other rights.
+
+       1. Moral rights, such as the right of integrity, are not
+          licensed under this Public License, nor are publicity,
+          privacy, and/or other similar personality rights; however, to
+          the extent possible, the Licensor waives and/or agrees not to
+          assert any such rights held by the Licensor to the limited
+          extent necessary to allow You to exercise the Licensed
+          Rights, but not otherwise.
+
+       2. Patent and trademark rights are not licensed under this
+          Public License.
+
+       3. To the extent possible, the Licensor waives any right to
+          collect royalties from You for the exercise of the Licensed
+          Rights, whether directly or through a collecting society
+          under any voluntary or waivable statutory or compulsory
+          licensing scheme. In all other cases the Licensor expressly
+          reserves any right to collect such royalties.
+
+
+Section 3 -- License Conditions.
+
+Your exercise of the Licensed Rights is expressly made subject to the
+following conditions.
+
+  a. Attribution.
+
+       1. If You Share the Licensed Material (including in modified
+          form), You must:
+
+            a. retain the following if it is supplied by the Licensor
+               with the Licensed Material:
+
+                 i. identification of the creator(s) of the Licensed
+                    Material and any others designated to receive
+                    attribution, in any reasonable manner requested by
+                    the Licensor (including by pseudonym if
+                    designated);
+
+                ii. a copyright notice;
+
+               iii. a notice that refers to this Public License;
+
+                iv. a notice that refers to the disclaimer of
+                    warranties;
+
+                 v. a URI or hyperlink to the Licensed Material to the
+                    extent reasonably practicable;
+
+            b. indicate if You modified the Licensed Material and
+               retain an indication of any previous modifications; and
+
+            c. indicate the Licensed Material is licensed under this
+               Public License, and include the text of, or the URI or
+               hyperlink to, this Public License.
+
+       2. You may satisfy the conditions in Section 3(a)(1) in any
+          reasonable manner based on the medium, means, and context in
+          which You Share the Licensed Material. For example, it may be
+          reasonable to satisfy the conditions by providing a URI or
+          hyperlink to a resource that includes the required
+          information.
+
+       3. If requested by the Licensor, You must remove any of the
+          information required by Section 3(a)(1)(A) to the extent
+          reasonably practicable.
+
+       4. If You Share Adapted Material You produce, the Adapter's
+          License You apply must not prevent recipients of the Adapted
+          Material from complying with this Public License.
+
+
+Section 4 -- Sui Generis Database Rights.
+
+Where the Licensed Rights include Sui Generis Database Rights that
+apply to Your use of the Licensed Material:
+
+  a. for the avoidance of doubt, Section 2(a)(1) grants You the right
+     to extract, reuse, reproduce, and Share all or a substantial
+     portion of the contents of the database;
+
+  b. if You include all or a substantial portion of the database
+     contents in a database in which You have Sui Generis Database
+     Rights, then the database in which You have Sui Generis Database
+     Rights (but not its individual contents) is Adapted Material; and
+
+  c. You must comply with the conditions in Section 3(a) if You Share
+     all or a substantial portion of the contents of the database.
+
+For the avoidance of doubt, this Section 4 supplements and does not
+replace Your obligations under this Public License where the Licensed
+Rights include other Copyright and Similar Rights.
+
+
+Section 5 -- Disclaimer of Warranties and Limitation of Liability.
+
+  a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
+     EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
+     AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
+     ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
+     IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
+     WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
+     PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
+     ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
+     KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
+     ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
+
+  b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
+     TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
+     NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
+     INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
+     COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
+     USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
+     ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
+     DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
+     IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
+
+  c. The disclaimer of warranties and limitation of liability provided
+     above shall be interpreted in a manner that, to the extent
+     possible, most closely approximates an absolute disclaimer and
+     waiver of all liability.
+
+
+Section 6 -- Term and Termination.
+
+  a. This Public License applies for the term of the Copyright and
+     Similar Rights licensed here. However, if You fail to comply with
+     this Public License, then Your rights under this Public License
+     terminate automatically.
+
+  b. Where Your right to use the Licensed Material has terminated under
+     Section 6(a), it reinstates:
+
+       1. automatically as of the date the violation is cured, provided
+          it is cured within 30 days of Your discovery of the
+          violation; or
+
+       2. upon express reinstatement by the Licensor.
+
+     For the avoidance of doubt, this Section 6(b) does not affect any
+     right the Licensor may have to seek remedies for Your violations
+     of this Public License.
+
+  c. For the avoidance of doubt, the Licensor may also offer the
+     Licensed Material under separate terms or conditions or stop
+     distributing the Licensed Material at any time; however, doing so
+     will not terminate this Public License.
+
+  d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
+     License.
+
+
+Section 7 -- Other Terms and Conditions.
+
+  a. The Licensor shall not be bound by any additional or different
+     terms or conditions communicated by You unless expressly agreed.
+
+  b. Any arrangements, understandings, or agreements regarding the
+     Licensed Material not stated herein are separate from and
+     independent of the terms and conditions of this Public License.
+
+
+Section 8 -- Interpretation.
+
+  a. For the avoidance of doubt, this Public License does not, and
+     shall not be interpreted to, reduce, limit, restrict, or impose
+     conditions on any use of the Licensed Material that could lawfully
+     be made without permission under this Public License.
+
+  b. To the extent possible, if any provision of this Public License is
+     deemed unenforceable, it shall be automatically reformed to the
+     minimum extent necessary to make it enforceable. If the provision
+     cannot be reformed, it shall be severed from this Public License
+     without affecting the enforceability of the remaining terms and
+     conditions.
+
+  c. No term or condition of this Public License will be waived and no
+     failure to comply consented to unless expressly agreed to by the
+     Licensor.
+
+  d. Nothing in this Public License constitutes or may be interpreted
+     as a limitation upon, or waiver of, any privileges and immunities
+     that apply to the Licensor or You, including from the legal
+     processes of any jurisdiction or authority.
+
+
+=======================================================================
+
+Creative Commons is not a party to its public licenses.
+Notwithstanding, Creative Commons may elect to apply one of its public
+licenses to material it publishes and in those instances will be
+considered the "Licensor." Except for the limited purpose of indicating
+that material is shared under a Creative Commons public license or as
+otherwise permitted by the Creative Commons policies published at
+creativecommons.org/policies, Creative Commons does not authorize the
+use of the trademark "Creative Commons" or any other trademark or logo
+of Creative Commons without its prior written consent including,
+without limitation, in connection with any unauthorized modifications
+to any of its public licenses or any other arrangements,
+understandings, or agreements concerning use of licensed material. For
+the avoidance of doubt, this paragraph does not form part of the public
+licenses.
+
+Creative Commons may be contacted at creativecommons.org.

src/ndl_layout/README.md

@@ -0,0 +1,92 @@
+# NDLOCR用レイアウト認識モジュール
+
+レイアウト要素を抽出するためのモジュールのリポジトリです。
+
+本プログラムは、国立国会図書館が株式会社モルフォAIソリューションズに委託して作成したものです。
+
+本プログラムは、国立国会図書館がCC BY 4.0ライセンスで公開するものです。詳細については
+[LICENSE](./LICENSE
+)をご覧ください。
+
+# 環境構築
+
+python3.7かつ、cuda 11.1をインストール済みの環境の場合
+ndl_layoutディレクトリ直下で以下のコマンドを実行する。
+```
+pip install torch==1.8.1+cu111 torchvision==0.9.1+cu111 -f https://download.pytorch.org/whl/lts/1.8/torch_lts.html
+wget https://lab.ndl.go.jp/dataset/ndlocr/ndl_layout/ndl_layout_config.py -P ./models
+wget https://lab.ndl.go.jp/dataset/ndlocr/ndl_layout/epoch_140_all_eql_bt.pth -P ./models
+```
+
+くわえて、元リポジトリ(https://github.com/open-mmlab/mmdetection)
+をカスタマイズした[mmdetection](https://github.com/ndl-lab/mmdetection)
+に依存しているため、下記のようにリポジトリの追加とインストールを行う。
+
+```bash
+git clone https://github.com/ndl-lab/mmdetection
+cd mmdetection
+python setup.py bdist_wheel
+pip install dist/*.whl
+```
+
+
+# 使い方
+※スクリプトファイルはndl_layoutディレクトリ直下で実行すること
+
+## tools/process.py : 推論用モジュール + CLI
+
+学習結果を使って推論を実行する。学習済みのモデルは`ndl_layout/models` 以下にあるものとする。
+
+画像リストを引数で指定するには img_paths オプションを、画像リストをファイルから読み込む場合には list_path オプションを指定する。
+
+output_path で出力 XML ファイルの格納先を変更することができる。（デフォルトは layout_prediction.xml）
+
+use_show オプションを追加すると処理結果をGUI上で確認することができる。
+
+img_pathsオプションで画像リストを指定する例
+```bash
+python -m tools.process --img_paths image/dir/path/*.jpg --use_show --output_path layout_prediction.xml --config ./models/ndl_layout_config.py --checkpoint ./models/epoch_140_all_eql_bt.pth
+```
+
+list_path オプションで画像リストを指定する例
+```bash
+python -m tools.process --list_path image_list_file.list --use_show --output_path layout_prediction.xml --config ./models/ndl_layout_config.py --checkpoint ./models/epoch_140_all_eql_bt.pth
+```
+
+## tools/preprocess.py : 学習画像の追加＆変換
+
+画像のファイル名の変換、縮小を行い、MS COCO 形式に整形。
+
+```bash
+python -m tools.preprocess images_data_dir output_dir --use_link
+```
+
+出力解像度を下げる必要がない場合には、`--use_link`オプションを指定する。
+
+高解像の場合など、解像度を下げたい場合には `--use_shrink` を使うと画像サイズとアノテーションを半分のサイズに縮小して出力する。
+
+本リポジトリの追加学習に使用可能なファイル(アノテーション情報の含まれるjson及び、前処理後の画像)は `output_dir` で指定したディレクトリに出力される。 
+
+
+## 学習時の手順
+1) ndl_layout/tools/preprocess.pyを使用し、NDLOCRXMLDataset形式の画像とアノテーションファイル(xml)をCOCO形式に変換し保存する。
+```
+cd mmdetection
+python -m tools.preprocess images_data_dir output_dir --use_link
+```
+output_dir内に画像のシンボリックリンク（またはコピー）とCOCO形式のアノテーションファイル(.json)を保存する。
+
+アノテーションファイルは、data.json(全データのアノテーション)、train.json(ランダムに全体の9割)、test.json(train以外の残る1割)を生成する。
+
+2) mmdetection/tools/train_ndl.py を使用し、モデルを学習する。
+```
+cd mmdetection
+python tools/train_ndl.py configs/ndl/cascade_rcnn_r50_fpn_1x_ndl_1024_eql.py
+```
+学習データ、work directory、初期値、学習回数等はconfigファイル内で指定するか、train_ndl.pyのオプションを使用する。オプションで指定されたものが優先される。
+
+work directoryに、学習したモデル(epoch_XX.pth または latest.pth)とconfigファイル(train_ndl.pyのオプションを使用した場合その内容も反映)、学習時のログファイル(.logと.log.json)が保存される。
+
+なお、このリポジトリで公開しているモデル（設定ファイルは`configs/ndl/cascade_rcnn_r50_fpn_1x_ndl_1024_eql.py`を参照）の学習時の初期重みには
+https://download.openmmlab.com/mmdetection/v2.0/cascade_rcnn/cascade_rcnn_r50_fpn_1x_coco/cascade_rcnn_r50_fpn_1x_coco_20200316-3dc56deb.pth
+を使用した。

src/ndl_layout/mmdetection/.dev_scripts/batch_test.py

@@ -0,0 +1,212 @@
+"""
+some instructions
+1. Fill the models that needs to be checked in the modelzoo_dict
+2. Arange the structure of the directory as follows, the script will find the
+   corresponding config itself:
+   model_dir/model_family/checkpoints
+   e.g.: models/faster_rcnn/faster_rcnn_r50_fpn_1x_coco_20200130-047c8118.pth
+         models/faster_rcnn/faster_rcnn_r101_fpn_1x_coco_20200130-047c8118.pth
+3. Excute the batch_test.sh
+"""
+
+import argparse
+import json
+import os
+import subprocess
+
+import mmcv
+import torch
+from mmcv import Config, get_logger
+from mmcv.parallel import MMDataParallel, MMDistributedDataParallel
+from mmcv.runner import (get_dist_info, init_dist, load_checkpoint,
+                         wrap_fp16_model)
+
+from mmdet.apis import multi_gpu_test, single_gpu_test
+from mmdet.datasets import (build_dataloader, build_dataset,
+                            replace_ImageToTensor)
+from mmdet.models import build_detector
+
+modelzoo_dict = {
+    'configs/faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py': {
+        'bbox': 0.374
+    },
+    'configs/mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py': {
+        'bbox': 0.382,
+        'segm': 0.347
+    },
+    'configs/rpn/rpn_r50_fpn_1x_coco.py': {
+        'AR@1000': 0.582
+    }
+}
+
+
+def parse_args():
+    parser = argparse.ArgumentParser(
+        description='The script used for checking the correctness \
+            of batch inference')
+    parser.add_argument('model_dir', help='directory of models')
+    parser.add_argument(
+        'json_out', help='the output json records test information like mAP')
+    parser.add_argument(
+        '--launcher',
+        choices=['none', 'pytorch', 'slurm', 'mpi'],
+        default='none',
+        help='job launcher')
+    parser.add_argument('--local_rank', type=int, default=0)
+    args = parser.parse_args()
+    if 'LOCAL_RANK' not in os.environ:
+        os.environ['LOCAL_RANK'] = str(args.local_rank)
+    return args
+
+
+def check_finish(all_model_dict, result_file):
+    # check if all models are checked
+    tested_cfgs = []
+    with open(result_file, 'r+') as f:
+        for line in f:
+            line = json.loads(line)
+            tested_cfgs.append(line['cfg'])
+    is_finish = True
+    for cfg in sorted(all_model_dict.keys()):
+        if cfg not in tested_cfgs:
+            return cfg
+    if is_finish:
+        with open(result_file, 'a+') as f:
+            f.write('finished\n')
+
+
+def dump_dict(record_dict, json_out):
+    # dump result json dict
+    with open(json_out, 'a+') as f:
+        mmcv.dump(record_dict, f, file_format='json')
+        f.write('\n')
+
+
+def main():
+    args = parse_args()
+    # touch the output json if not exist
+    with open(args.json_out, 'a+'):
+        pass
+    # init distributed env first, since logger depends on the dist
+    # info.
+    if args.launcher == 'none':
+        distributed = False
+    else:
+        distributed = True
+        init_dist(args.launcher, backend='nccl')
+    rank, world_size = get_dist_info()
+
+    logger = get_logger('root')
+
+    # read info of checkpoints and config
+    result_dict = dict()
+    for model_family_dir in os.listdir(args.model_dir):
+        for model in os.listdir(
+                os.path.join(args.model_dir, model_family_dir)):
+            # cpt: rpn_r50_fpn_1x_coco_20200218-5525fa2e.pth
+            # cfg: rpn_r50_fpn_1x_coco.py
+            cfg = model.split('.')[0][:-18] + '.py'
+            cfg_path = os.path.join('configs', model_family_dir, cfg)
+            assert os.path.isfile(
+                cfg_path), f'{cfg_path} is not valid config path'
+            cpt_path = os.path.join(args.model_dir, model_family_dir, model)
+            result_dict[cfg_path] = cpt_path
+            assert cfg_path in modelzoo_dict, f'please fill the ' \
+                                              f'performance of cfg: {cfg_path}'
+    cfg = check_finish(result_dict, args.json_out)
+    cpt = result_dict[cfg]
+    try:
+        cfg_name = cfg
+        logger.info(f'evaluate {cfg}')
+        record = dict(cfg=cfg, cpt=cpt)
+        cfg = Config.fromfile(cfg)
+        # cfg.data.test.ann_file = 'data/val_0_10.json'
+        # set cudnn_benchmark
+        if cfg.get('cudnn_benchmark', False):
+            torch.backends.cudnn.benchmark = True
+        cfg.model.pretrained = None
+        if cfg.model.get('neck'):
+            if isinstance(cfg.model.neck, list):
+                for neck_cfg in cfg.model.neck:
+                    if neck_cfg.get('rfp_backbone'):
+                        if neck_cfg.rfp_backbone.get('pretrained'):
+                            neck_cfg.rfp_backbone.pretrained = None
+            elif cfg.model.neck.get('rfp_backbone'):
+                if cfg.model.neck.rfp_backbone.get('pretrained'):
+                    cfg.model.neck.rfp_backbone.pretrained = None
+
+        # in case the test dataset is concatenated
+        if isinstance(cfg.data.test, dict):
+            cfg.data.test.test_mode = True
+        elif isinstance(cfg.data.test, list):
+            for ds_cfg in cfg.data.test:
+                ds_cfg.test_mode = True
+
+        # build the dataloader
+        samples_per_gpu = 2  # hack test with 2 image per gpu
+        if samples_per_gpu > 1:
+            # Replace 'ImageToTensor' to 'DefaultFormatBundle'
+            cfg.data.test.pipeline = replace_ImageToTensor(
+                cfg.data.test.pipeline)
+        dataset = build_dataset(cfg.data.test)
+        data_loader = build_dataloader(
+            dataset,
+            samples_per_gpu=samples_per_gpu,
+            workers_per_gpu=cfg.data.workers_per_gpu,
+            dist=distributed,
+            shuffle=False)
+
+        # build the model and load checkpoint
+        cfg.model.train_cfg = None
+        model = build_detector(cfg.model, test_cfg=cfg.get('test_cfg'))
+        fp16_cfg = cfg.get('fp16', None)
+        if fp16_cfg is not None:
+            wrap_fp16_model(model)
+
+        checkpoint = load_checkpoint(model, cpt, map_location='cpu')
+        # old versions did not save class info in checkpoints,
+        # this walkaround is for backward compatibility
+        if 'CLASSES' in checkpoint.get('meta', {}):
+            model.CLASSES = checkpoint['meta']['CLASSES']
+        else:
+            model.CLASSES = dataset.CLASSES
+
+        if not distributed:
+            model = MMDataParallel(model, device_ids=[0])
+            outputs = single_gpu_test(model, data_loader)
+        else:
+            model = MMDistributedDataParallel(
+                model.cuda(),
+                device_ids=[torch.cuda.current_device()],
+                broadcast_buffers=False)
+            outputs = multi_gpu_test(model, data_loader, 'tmp')
+        if rank == 0:
+            ref_mAP_dict = modelzoo_dict[cfg_name]
+            metrics = list(ref_mAP_dict.keys())
+            metrics = [
+                m if m != 'AR@1000' else 'proposal_fast' for m in metrics
+            ]
+            eval_results = dataset.evaluate(outputs, metrics)
+            print(eval_results)
+            for metric in metrics:
+                if metric == 'proposal_fast':
+                    ref_metric = modelzoo_dict[cfg_name]['AR@1000']
+                    eval_metric = eval_results['AR@1000']
+                else:
+                    ref_metric = modelzoo_dict[cfg_name][metric]
+                    eval_metric = eval_results[f'{metric}_mAP']
+                if abs(ref_metric - eval_metric) > 0.003:
+                    record['is_normal'] = False
+            dump_dict(record, args.json_out)
+            check_finish(result_dict, args.json_out)
+    except Exception as e:
+        logger.error(f'rank: {rank} test fail with error: {e}')
+        record['terminate'] = True
+        dump_dict(record, args.json_out)
+        check_finish(result_dict, args.json_out)
+        # hack there to throw some error to prevent hang out
+        subprocess.call('xxx')
+
+
+if __name__ == '__main__':
+    main()

src/ndl_layout/mmdetection/.dev_scripts/batch_test.sh

@@ -0,0 +1,19 @@
+export PYTHONPATH=${PWD}
+
+partition=$1
+model_dir=$2
+json_out=$3
+job_name=batch_test
+gpus=8
+gpu_per_node=8
+
+touch $json_out
+lastLine=$(tail -n 1 $json_out)
+while [ "$lastLine" != "finished" ]
+do
+    srun -p ${partition} --gres=gpu:${gpu_per_node} -n${gpus} --ntasks-per-node=${gpu_per_node} \
+        --job-name=${job_name} --kill-on-bad-exit=1 \
+        python .dev_scripts/batch_test.py $model_dir $json_out --launcher='slurm'
+    lastLine=$(tail -n 1 $json_out)
+    echo $lastLine
+done

src/ndl_layout/mmdetection/.dev_scripts/benchmark_filter.py

@@ -0,0 +1,158 @@
+import argparse
+import os
+import os.path as osp
+
+import mmcv
+
+
+def parse_args():
+    parser = argparse.ArgumentParser(description='Filter configs to train')
+    parser.add_argument(
+        '--basic-arch',
+        action='store_true',
+        help='to train models in basic arch')
+    parser.add_argument(
+        '--datasets', action='store_true', help='to train models in dataset')
+    parser.add_argument(
+        '--data-pipeline',
+        action='store_true',
+        help='to train models related to data pipeline, e.g. augmentations')
+    parser.add_argument(
+        '--nn-module',
+        action='store_true',
+        help='to train models related to neural network modules')
+    parser.add_argument(
+        '--model-options',
+        nargs='+',
+        help='custom options to special model benchmark')
+
+    args = parser.parse_args()
+    return args
+
+
+basic_arch_root = [
+    'atss', 'cascade_rcnn', 'cascade_rpn', 'centripetalnet', 'cornernet',
+    'detectors', 'detr', 'double_heads', 'dynamic_rcnn', 'faster_rcnn', 'fcos',
+    'foveabox', 'fp16', 'free_anchor', 'fsaf', 'gfl', 'ghm', 'grid_rcnn',
+    'guided_anchoring', 'htc', 'libra_rcnn', 'mask_rcnn', 'ms_rcnn',
+    'nas_fcos', 'paa', 'pisa', 'point_rend', 'reppoints', 'retinanet', 'rpn',
+    'sabl', 'ssd', 'tridentnet', 'vfnet', 'yolact', 'yolo', 'sparse_rcnn',
+    'scnet'
+]
+
+datasets_root = [
+    'wider_face', 'pascal_voc', 'cityscapes', 'lvis', 'deepfashion'
+]
+
+data_pipeline_root = ['albu_example', 'instaboost']
+
+nn_module_root = [
+    'carafe', 'dcn', 'empirical_attention', 'gcnet', 'gn', 'gn+ws', 'hrnet',
+    'pafpn', 'nas_fpn', 'regnet', 'resnest', 'res2net', 'groie'
+]
+
+benchmark_pool = [
+    'configs/albu_example/mask_rcnn_r50_fpn_albu_1x_coco.py',
+    'configs/atss/atss_r50_fpn_1x_coco.py',
+    'configs/carafe/mask_rcnn_r50_fpn_carafe_1x_coco.py',
+    'configs/cascade_rcnn/cascade_mask_rcnn_r50_fpn_1x_coco.py',
+    'configs/cascade_rpn/crpn_faster_rcnn_r50_caffe_fpn_1x_coco.py',
+    'configs/centripetalnet/'
+    'centripetalnet_hourglass104_mstest_16x6_210e_coco.py',
+    'configs/cityscapes/mask_rcnn_r50_fpn_1x_cityscapes.py',
+    'configs/cornernet/'
+    'cornernet_hourglass104_mstest_8x6_210e_coco.py',  # special
+    'configs/dcn/mask_rcnn_r50_fpn_mdconv_c3-c5_1x_coco.py',
+    'configs/dcn/faster_rcnn_r50_fpn_dpool_1x_coco.py',
+    'configs/dcn/faster_rcnn_r50_fpn_mdpool_1x_coco.py',
+    'configs/dcn/mask_rcnn_r50_fpn_dconv_c3-c5_1x_coco.py',
+    'configs/detectors/detectors_htc_r50_1x_coco.py',
+    'configs/detr/detr_r50_8x2_150e_coco.py',
+    'configs/double_heads/dh_faster_rcnn_r50_fpn_1x_coco.py',
+    'configs/dynamic_rcnn/dynamic_rcnn_r50_fpn_1x.py',
+    'configs/empirical_attention/faster_rcnn_r50_fpn_attention_1111_dcn_1x_coco.py',  # noqa
+    'configs/faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py',
+    'configs/faster_rcnn/faster_rcnn_r50_fpn_ohem_1x_coco.py',
+    'configs/faster_rcnn/faster_rcnn_r50_caffe_fpn_1x_coco.py',
+    'configs/faster_rcnn/faster_rcnn_r50_caffe_fpn_mstrain_1x_coco.py',
+    'configs/faster_rcnn/faster_rcnn_r50_caffe_dc5_mstrain_1x_coco.py',
+    'configs/fcos/fcos_center_r50_caffe_fpn_gn-head_4x4_1x_coco.py',
+    'configs/foveabox/fovea_align_r50_fpn_gn-head_4x4_2x_coco.py',
+    'configs/fp16/retinanet_r50_fpn_fp16_1x_coco.py',
+    'configs/fp16/mask_rcnn_r50_fpn_fp16_1x_coco.py',
+    'configs/free_anchor/retinanet_free_anchor_r50_fpn_1x_coco.py',
+    'configs/fsaf/fsaf_r50_fpn_1x_coco.py',
+    'configs/gcnet/mask_rcnn_r50_fpn_r4_gcb_c3-c5_1x_coco.py',
+    'configs/gfl/gfl_r50_fpn_1x_coco.py',
+    'configs/ghm/retinanet_ghm_r50_fpn_1x_coco.py',
+    'configs/gn/mask_rcnn_r50_fpn_gn-all_2x_coco.py',
+    'configs/gn+ws/mask_rcnn_r50_fpn_gn_ws-all_2x_coco.py',
+    'configs/grid_rcnn/grid_rcnn_r50_fpn_gn-head_2x_coco.py',
+    'configs/groie/faster_rcnn_r50_fpn_groie_1x_coco.py',
+    'configs/guided_anchoring/ga_faster_r50_caffe_fpn_1x_coco.py',
+    'configs/hrnet/mask_rcnn_hrnetv2p_w18_1x_coco.py',
+    'configs/htc/htc_r50_fpn_1x_coco.py',
+    'configs/instaboost/mask_rcnn_r50_fpn_instaboost_4x_coco.py',
+    'configs/libra_rcnn/libra_faster_rcnn_r50_fpn_1x_coco.py',
+    'configs/lvis/mask_rcnn_r50_fpn_sample1e-3_mstrain_1x_lvis_v1.py',
+    'configs/mask_rcnn/mask_rcnn_r50_caffe_fpn_mstrain-poly_1x_coco.py',
+    'configs/ms_rcnn/ms_rcnn_r50_caffe_fpn_1x_coco.py',
+    'configs/nas_fcos/nas_fcos_nashead_r50_caffe_fpn_gn-head_4x4_1x_coco.py',
+    'configs/nas_fpn/retinanet_r50_nasfpn_crop640_50e_coco.py',
+    'configs/paa/paa_r50_fpn_1x_coco.py',
+    'configs/pafpn/faster_rcnn_r50_pafpn_1x_coco.py',
+    'configs/pisa/pisa_mask_rcnn_r50_fpn_1x_coco.py',
+    'configs/point_rend/point_rend_r50_caffe_fpn_mstrain_1x_coco.py',
+    'configs/regnet/mask_rcnn_regnetx-3.2GF_fpn_1x_coco.py',
+    'configs/reppoints/reppoints_moment_r50_fpn_gn-neck+head_1x_coco.py',
+    'configs/res2net/faster_rcnn_r2_101_fpn_2x_coco.py',
+    'configs/resnest/'
+    'mask_rcnn_s50_fpn_syncbn-backbone+head_mstrain_1x_coco.py',
+    'configs/retinanet/retinanet_r50_caffe_fpn_1x_coco.py',
+    'configs/rpn/rpn_r50_fpn_1x_coco.py',
+    'configs/sabl/sabl_retinanet_r50_fpn_1x_coco.py',
+    'configs/ssd/ssd300_coco.py',
+    'configs/tridentnet/tridentnet_r50_caffe_1x_coco.py',
+    'configs/vfnet/vfnet_r50_fpn_1x_coco.py',
+    'configs/yolact/yolact_r50_1x8_coco.py',
+    'configs/yolo/yolov3_d53_320_273e_coco.py',
+    'configs/sparse_rcnn/sparse_rcnn_r50_fpn_1x_coco.py',
+    'configs/scnet/scnet_r50_fpn_1x_coco.py'
+]
+
+
+def main():
+    args = parse_args()
+
+    benchmark_type = []
+    if args.basic_arch:
+        benchmark_type += basic_arch_root
+    if args.datasets:
+        benchmark_type += datasets_root
+    if args.data_pipeline:
+        benchmark_type += data_pipeline_root
+    if args.nn_module:
+        benchmark_type += nn_module_root
+
+    special_model = args.model_options
+    if special_model is not None:
+        benchmark_type += special_model
+
+    config_dpath = 'configs/'
+    benchmark_configs = []
+    for cfg_root in benchmark_type:
+        cfg_dir = osp.join(config_dpath, cfg_root)
+        configs = os.scandir(cfg_dir)
+        for cfg in configs:
+            config_path = osp.join(cfg_dir, cfg.name)
+            if (config_path in benchmark_pool
+                    and config_path not in benchmark_configs):
+                benchmark_configs.append(config_path)
+
+    print(f'Totally found {len(benchmark_configs)} configs to benchmark')
+    config_dicts = dict(models=benchmark_configs)
+    mmcv.dump(config_dicts, 'regression_test_configs.json')
+
+
+if __name__ == '__main__':
+    main()

src/ndl_layout/mmdetection/.dev_scripts/convert_benchmark_script.py

@@ -0,0 +1,86 @@
+import argparse
+import os
+import os.path as osp
+
+import mmcv
+
+
+def parse_args():
+    parser = argparse.ArgumentParser(
+        description='Convert benchmark model json to script')
+    parser.add_argument(
+        'json_path', type=str, help='json path output by benchmark_filter')
+    parser.add_argument('partition', type=str, help='slurm partition name')
+    parser.add_argument(
+        '--max-keep-ckpts',
+        type=int,
+        default=1,
+        help='The maximum checkpoints to keep')
+    parser.add_argument(
+        '--run', action='store_true', help='run script directly')
+    parser.add_argument(
+        '--out', type=str, help='path to save model benchmark script')
+
+    args = parser.parse_args()
+    return args
+
+
+def main():
+    args = parse_args()
+    if args.out:
+        out_suffix = args.out.split('.')[-1]
+        assert args.out.endswith('.sh'), \
+            f'Expected out file path suffix is .sh, but get .{out_suffix}'
+    assert args.out or args.run, \
+        ('Please specify at least one operation (save/run/ the '
+         'script) with the argument "--out" or "--run"')
+
+    json_data = mmcv.load(args.json_path)
+    model_cfgs = json_data['models']
+
+    partition = args.partition  # cluster name
+
+    root_name = './tools'
+    train_script_name = osp.join(root_name, 'slurm_train.sh')
+    # stdout is no output
+    stdout_cfg = '>/dev/null'
+
+    max_keep_ckpts = args.max_keep_ckpts
+
+    commands = []
+    for i, cfg in enumerate(model_cfgs):
+        # print cfg name
+        echo_info = f'echo \'{cfg}\' &'
+        commands.append(echo_info)
+        commands.append('\n')
+
+        fname, _ = osp.splitext(osp.basename(cfg))
+        out_fname = osp.join(root_name, fname)
+        # default setting
+        command_info = f'GPUS=8  GPUS_PER_NODE=8  ' \
+                       f'CPUS_PER_TASK=2 {train_script_name} '
+        command_info += f'{partition} '
+        command_info += f'{fname} '
+        command_info += f'{cfg} '
+        command_info += f'{out_fname} '
+        if max_keep_ckpts:
+            command_info += f'--cfg-options ' \
+                            f'checkpoint_config.max_keep_ckpts=' \
+                            f'{max_keep_ckpts}' + ' '
+        command_info += f'{stdout_cfg} &'
+
+        commands.append(command_info)
+
+        if i < len(model_cfgs):
+            commands.append('\n')
+
+    command_str = ''.join(commands)
+    if args.out:
+        with open(args.out, 'w') as f:
+            f.write(command_str)
+    if args.run:
+        os.system(command_str)
+
+
+if __name__ == '__main__':
+    main()

src/ndl_layout/mmdetection/.dev_scripts/gather_benchmark_metric.py

@@ -0,0 +1,142 @@
+import argparse
+import glob
+import os.path as osp
+
+import mmcv
+from gather_models import get_final_results
+
+try:
+    import xlrd
+except ImportError:
+    xlrd = None
+try:
+    import xlutils
+    from xlutils.copy import copy
+except ImportError:
+    xlutils = None
+
+
+def parse_args():
+    parser = argparse.ArgumentParser(
+        description='Gather benchmarked models metric')
+    parser.add_argument(
+        'root',
+        type=str,
+        help='root path of benchmarked models to be gathered')
+    parser.add_argument(
+        'benchmark_json', type=str, help='json path of benchmark models')
+    parser.add_argument(
+        '--out', type=str, help='output path of gathered metrics to be stored')
+    parser.add_argument(
+        '--not-show', action='store_true', help='not show metrics')
+    parser.add_argument(
+        '--excel', type=str, help='input path of excel to be recorded')
+    parser.add_argument(
+        '--ncol', type=int, help='Number of column to be modified or appended')
+
+    args = parser.parse_args()
+    return args
+
+
+if __name__ == '__main__':
+    args = parse_args()
+
+    if args.excel:
+        assert args.ncol, 'Please specify "--excel" and "--ncol" ' \
+                          'at the same time'
+        if xlrd is None:
+            raise RuntimeError(
+                'xlrd is not installed,'
+                'Please use “pip install xlrd==1.2.0” to install')
+        if xlutils is None:
+            raise RuntimeError(
+                'xlutils is not installed,'
+                'Please use “pip install xlutils==2.0.0” to install')
+        readbook = xlrd.open_workbook(args.excel)
+        sheet = readbook.sheet_by_name('Sheet1')
+        sheet_info = {}
+        total_nrows = sheet.nrows
+        for i in range(3, sheet.nrows):
+            sheet_info[sheet.row_values(i)[0]] = i
+        xlrw = copy(readbook)
+        table = xlrw.get_sheet(0)
+
+    root_path = args.root
+    metrics_out = args.out
+    benchmark_json_path = args.benchmark_json
+    model_configs = mmcv.load(benchmark_json_path)['models']
+
+    result_dict = {}
+    for config in model_configs:
+        config_name = osp.split(config)[-1]
+        config_name = osp.splitext(config_name)[0]
+        result_path = osp.join(root_path, config_name)
+        if osp.exists(result_path):
+            # 1 read config
+            cfg = mmcv.Config.fromfile(config)
+            total_epochs = cfg.runner.max_epochs
+            final_results = cfg.evaluation.metric
+            if not isinstance(final_results, list):
+                final_results = [final_results]
+            final_results_out = []
+            for key in final_results:
+                if 'proposal_fast' in key:
+                    final_results_out.append('AR@1000')  # RPN
+                elif 'mAP' not in key:
+                    final_results_out.append(key + '_mAP')
+
+            # 2 determine whether total_epochs ckpt exists
+            ckpt_path = f'epoch_{total_epochs}.pth'
+            if osp.exists(osp.join(result_path, ckpt_path)):
+                log_json_path = list(
+                    sorted(glob.glob(osp.join(result_path, '*.log.json'))))[-1]
+
+                # 3 read metric
+                model_performance = get_final_results(log_json_path,
+                                                      total_epochs,
+                                                      final_results_out)
+                if model_performance is None:
+                    print(f'log file error: {log_json_path}')
+                    continue
+                for performance in model_performance:
+                    if performance in ['AR@1000', 'bbox_mAP', 'segm_mAP']:
+                        metric = round(model_performance[performance] * 100, 1)
+                        model_performance[performance] = metric
+                result_dict[config] = model_performance
+
+                # update and append excel content
+                if args.excel:
+                    if 'AR@1000' in model_performance:
+                        metrics = f'{model_performance["AR@1000"]}(AR@1000)'
+                    elif 'segm_mAP' in model_performance:
+                        metrics = f'{model_performance["bbox_mAP"]}/' \
+                                  f'{model_performance["segm_mAP"]}'
+                    else:
+                        metrics = f'{model_performance["bbox_mAP"]}'
+
+                    row_num = sheet_info.get(config, None)
+                    if row_num:
+                        table.write(row_num, args.ncol, metrics)
+                    else:
+                        table.write(total_nrows, 0, config)
+                        table.write(total_nrows, args.ncol, metrics)
+                        total_nrows += 1
+
+            else:
+                print(f'{config} not exist: {ckpt_path}')
+        else:
+            print(f'not exist: {config}')
+
+    # 4 save or print results
+    if metrics_out:
+        mmcv.mkdir_or_exist(metrics_out)
+        mmcv.dump(result_dict, osp.join(metrics_out, 'model_metric_info.json'))
+    if not args.not_show:
+        print('===================================')
+        for config_name, metrics in result_dict.items():
+            print(config_name, metrics)
+        print('===================================')
+    if args.excel:
+        filename, sufflx = osp.splitext(args.excel)
+        xlrw.save(f'{filename}_o{sufflx}')
+        print(f'>>> Output {filename}_o{sufflx}')

src/ndl_layout/mmdetection/.dev_scripts/gather_models.py

@@ -0,0 +1,162 @@
+import argparse
+import glob
+import json
+import os.path as osp
+import shutil
+import subprocess
+
+import mmcv
+import torch
+
+
+def process_checkpoint(in_file, out_file):
+    checkpoint = torch.load(in_file, map_location='cpu')
+    # remove optimizer for smaller file size
+    if 'optimizer' in checkpoint:
+        del checkpoint['optimizer']
+    # if it is necessary to remove some sensitive data in checkpoint['meta'],
+    # add the code here.
+    torch.save(checkpoint, out_file)
+    sha = subprocess.check_output(['sha256sum', out_file]).decode()
+    final_file = out_file.rstrip('.pth') + '-{}.pth'.format(sha[:8])
+    subprocess.Popen(['mv', out_file, final_file])
+    return final_file
+
+
+def get_final_epoch(config):
+    cfg = mmcv.Config.fromfile('./configs/' + config)
+    return cfg.total_epochs
+
+
+def get_final_results(log_json_path, epoch, results_lut):
+    result_dict = dict()
+    with open(log_json_path, 'r') as f:
+        for line in f.readlines():
+            log_line = json.loads(line)
+            if 'mode' not in log_line.keys():
+                continue
+
+            if log_line['mode'] == 'train' and log_line['epoch'] == epoch:
+                result_dict['memory'] = log_line['memory']
+
+            if log_line['mode'] == 'val' and log_line['epoch'] == epoch:
+                result_dict.update({
+                    key: log_line[key]
+                    for key in results_lut if key in log_line
+                })
+                return result_dict
+
+
+def parse_args():
+    parser = argparse.ArgumentParser(description='Gather benchmarked models')
+    parser.add_argument(
+        'root',
+        type=str,
+        help='root path of benchmarked models to be gathered')
+    parser.add_argument(
+        'out', type=str, help='output path of gathered models to be stored')
+
+    args = parser.parse_args()
+    return args
+
+
+def main():
+    args = parse_args()
+    models_root = args.root
+    models_out = args.out
+    mmcv.mkdir_or_exist(models_out)
+
+    # find all models in the root directory to be gathered
+    raw_configs = list(mmcv.scandir('./configs', '.py', recursive=True))
+
+    # filter configs that is not trained in the experiments dir
+    used_configs = []
+    for raw_config in raw_configs:
+        if osp.exists(osp.join(models_root, raw_config)):
+            used_configs.append(raw_config)
+    print(f'Find {len(used_configs)} models to be gathered')
+
+    # find final_ckpt and log file for trained each config
+    # and parse the best performance
+    model_infos = []
+    for used_config in used_configs:
+        exp_dir = osp.join(models_root, used_config)
+        # check whether the exps is finished
+        final_epoch = get_final_epoch(used_config)
+        final_model = 'epoch_{}.pth'.format(final_epoch)
+        model_path = osp.join(exp_dir, final_model)
+
+        # skip if the model is still training
+        if not osp.exists(model_path):
+            continue
+
+        # get the latest logs
+        log_json_path = list(
+            sorted(glob.glob(osp.join(exp_dir, '*.log.json'))))[-1]
+        log_txt_path = list(sorted(glob.glob(osp.join(exp_dir, '*.log'))))[-1]
+        cfg = mmcv.Config.fromfile('./configs/' + used_config)
+        results_lut = cfg.evaluation.metric
+        if not isinstance(results_lut, list):
+            results_lut = [results_lut]
+        # case when using VOC, the evaluation key is only 'mAP'
+        results_lut = [key + '_mAP' for key in results_lut if 'mAP' not in key]
+        model_performance = get_final_results(log_json_path, final_epoch,
+                                              results_lut)
+
+        if model_performance is None:
+            continue
+
+        model_time = osp.split(log_txt_path)[-1].split('.')[0]
+        model_infos.append(
+            dict(
+                config=used_config,
+                results=model_performance,
+                epochs=final_epoch,
+                model_time=model_time,
+                log_json_path=osp.split(log_json_path)[-1]))
+
+    # publish model for each checkpoint
+    publish_model_infos = []
+    for model in model_infos:
+        model_publish_dir = osp.join(models_out, model['config'].rstrip('.py'))
+        mmcv.mkdir_or_exist(model_publish_dir)
+
+        model_name = osp.split(model['config'])[-1].split('.')[0]
+
+        model_name += '_' + model['model_time']
+        publish_model_path = osp.join(model_publish_dir, model_name)
+        trained_model_path = osp.join(models_root, model['config'],
+                                      'epoch_{}.pth'.format(model['epochs']))
+
+        # convert model
+        final_model_path = process_checkpoint(trained_model_path,
+                                              publish_model_path)
+
+        # copy log
+        shutil.copy(
+            osp.join(models_root, model['config'], model['log_json_path']),
+            osp.join(model_publish_dir, f'{model_name}.log.json'))
+        shutil.copy(
+            osp.join(models_root, model['config'],
+                     model['log_json_path'].rstrip('.json')),
+            osp.join(model_publish_dir, f'{model_name}.log'))
+
+        # copy config to guarantee reproducibility
+        config_path = model['config']
+        config_path = osp.join(
+            'configs',
+            config_path) if 'configs' not in config_path else config_path
+        target_cconfig_path = osp.split(config_path)[-1]
+        shutil.copy(config_path,
+                    osp.join(model_publish_dir, target_cconfig_path))
+
+        model['model_path'] = final_model_path
+        publish_model_infos.append(model)
+
+    models = dict(models=publish_model_infos)
+    print(f'Totally gathered {len(publish_model_infos)} models')
+    mmcv.dump(models, osp.join(models_out, 'model_info.json'))
+
+
+if __name__ == '__main__':
+    main()

src/ndl_layout/mmdetection/.dev_scripts/linter.sh

@@ -0,0 +1,3 @@
+yapf -r -i mmdet/ configs/ tests/ tools/
+isort -rc mmdet/ configs/ tests/ tools/
+flake8 .

src/ndl_layout/mmdetection/.gitignore

@@ -0,0 +1,121 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+.hypothesis/
+.pytest_cache/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# pyenv
+.python-version
+
+# celery beat schedule file
+celerybeat-schedule
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+
+data/
+data
+.vscode
+.idea
+.DS_Store
+
+# custom
+*.pkl
+*.pkl.json
+*.log.json
+work_dirs/
+
+# Pytorch
+*.pth
+*.py~
+*.sh~

src/ndl_layout/mmdetection/.pre-commit-config.yaml

@@ -0,0 +1,40 @@
+repos:
+  - repo: https://gitlab.com/pycqa/flake8.git
+    rev: 3.8.3
+    hooks:
+      - id: flake8
+  - repo: https://github.com/asottile/seed-isort-config
+    rev: v2.2.0
+    hooks:
+      - id: seed-isort-config
+  - repo: https://github.com/timothycrosley/isort
+    rev: 4.3.21
+    hooks:
+      - id: isort
+  - repo: https://github.com/pre-commit/mirrors-yapf
+    rev: v0.30.0
+    hooks:
+      - id: yapf
+  - repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v3.1.0
+    hooks:
+      - id: trailing-whitespace
+      - id: check-yaml
+      - id: end-of-file-fixer
+      - id: requirements-txt-fixer
+      - id: double-quote-string-fixer
+      - id: check-merge-conflict
+      - id: fix-encoding-pragma
+        args: ["--remove"]
+      - id: mixed-line-ending
+        args: ["--fix=lf"]
+  - repo: https://github.com/jumanjihouse/pre-commit-hooks
+    rev: 2.1.4
+    hooks:
+      - id: markdownlint
+        args: ["-r", "~MD002,~MD013,~MD024,~MD029,~MD033,~MD034,~MD036", "-t", "allow_different_nesting"]
+  - repo: https://github.com/myint/docformatter
+    rev: v1.3.1
+    hooks:
+      - id: docformatter
+        args: ["--in-place", "--wrap-descriptions", "79"]

src/ndl_layout/mmdetection/.readthedocs.yml

@@ -0,0 +1,7 @@
+version: 2
+
+python:
+    version: 3.7
+    install:
+      - requirements: requirements/docs.txt
+      - requirements: requirements/readthedocs.txt

src/ndl_layout/mmdetection/LICENSE

@@ -0,0 +1,643 @@
+This software is largely based on the following repository:
+
+https://github.com/open-mmlab/mmdetection
+
+The original software license is as follows.
+
+-----------------------------------------------------------------------
+
+Copyright 2018-2019 Open-MMLab. All rights reserved.
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2018-2019 Open-MMLab.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.
+
+
+-----------------------------------------------------------------------
+
+The following files have been modified from their original version:
+
+README.md, 
+configs/_base_/models/cascade_rcnn_r50_fpn.py,
+configs/_base_/schedules/schedule_1x.py,
+configs/ndl/cascade_mask_rcnn_r50_fpn_1x_ndl.py,
+configs/ndl/cascade_mask_rcnn_r50_fpn_1x_ndl_1024.py,
+configs/ndl/cascade_rcnn_r50_fpn_1x_ndl.py,
+configs/ndl/cascade_rcnn_r50_fpn_1x_ndl_1024.py,
+configs/ndl/cascade_rcnn_r50_fpn_1x_ndl_1024_eql.py,
+configs/ndl/ndl.py,
+configs/ndl/ndl_1024.py,
+configs/ndl/ndl_instance.py,
+configs/ndl/ndl_instance_1024.py,
+mmdet/core/post_processing/__init__.py,
+mmdet/core/post_processing/bbox_nms.py,
+mmdet/core/post_processing/merge_augs.py,
+mmdet/datasets/builder.py,
+mmdet/datasets/class_balance_dataset_wrapper.py,
+mmdet/datasets/coco.py,
+mmdet/datasets/max_iter_dataset_wrapper.py,
+mmdet/models/losses/__init__.py,
+mmdet/models/losses/eql.py,
+mmdet/models/losses/eqlv2.py,
+mmdet/models/losses/group_softmax.py,
+mmdet/utils/ndl_categories.py,
+tools/analysis_tools/coco_error_analysis.py,
+tools/train_ndl.py
+
+The following license applies for those modifications:
+
+-----------------------------------------------------------------------
+
+Copyright (c) 2022, National Diet Library, Japan.
+
+Attribution 4.0 International
+
+=======================================================================
+
+Creative Commons Corporation ("Creative Commons") is not a law firm and
+does not provide legal services or legal advice. Distribution of
+Creative Commons public licenses does not create a lawyer-client or
+other relationship. Creative Commons makes its licenses and related
+information available on an "as-is" basis. Creative Commons gives no
+warranties regarding its licenses, any material licensed under their
+terms and conditions, or any related information. Creative Commons
+disclaims all liability for damages resulting from their use to the
+fullest extent possible.
+
+Using Creative Commons Public Licenses
+
+Creative Commons public licenses provide a standard set of terms and
+conditions that creators and other rights holders may use to share
+original works of authorship and other material subject to copyright
+and certain other rights specified in the public license below. The
+following considerations are for informational purposes only, are not
+exhaustive, and do not form part of our licenses.
+
+     Considerations for licensors: Our public licenses are
+     intended for use by those authorized to give the public
+     permission to use material in ways otherwise restricted by
+     copyright and certain other rights. Our licenses are
+     irrevocable. Licensors should read and understand the terms
+     and conditions of the license they choose before applying it.
+     Licensors should also secure all rights necessary before
+     applying our licenses so that the public can reuse the
+     material as expected. Licensors should clearly mark any
+     material not subject to the license. This includes other CC-
+     licensed material, or material used under an exception or
+     limitation to copyright. More considerations for licensors:
+	wiki.creativecommons.org/Considerations_for_licensors
+
+     Considerations for the public: By using one of our public
+     licenses, a licensor grants the public permission to use the
+     licensed material under specified terms and conditions. If
+     the licensor's permission is not necessary for any reason--for
+     example, because of any applicable exception or limitation to
+     copyright--then that use is not regulated by the license. Our
+     licenses grant only permissions under copyright and certain
+     other rights that a licensor has authority to grant. Use of
+     the licensed material may still be restricted for other
+     reasons, including because others have copyright or other
+     rights in the material. A licensor may make special requests,
+     such as asking that all changes be marked or described.
+     Although not required by our licenses, you are encouraged to
+     respect those requests where reasonable. More_considerations
+     for the public: 
+	wiki.creativecommons.org/Considerations_for_licensees
+
+=======================================================================
+
+Creative Commons Attribution 4.0 International Public License
+
+By exercising the Licensed Rights (defined below), You accept and agree
+to be bound by the terms and conditions of this Creative Commons
+Attribution 4.0 International Public License ("Public License"). To the
+extent this Public License may be interpreted as a contract, You are
+granted the Licensed Rights in consideration of Your acceptance of
+these terms and conditions, and the Licensor grants You such rights in
+consideration of benefits the Licensor receives from making the
+Licensed Material available under these terms and conditions.
+
+
+Section 1 -- Definitions.
+
+  a. Adapted Material means material subject to Copyright and Similar
+     Rights that is derived from or based upon the Licensed Material
+     and in which the Licensed Material is translated, altered,
+     arranged, transformed, or otherwise modified in a manner requiring
+     permission under the Copyright and Similar Rights held by the
+     Licensor. For purposes of this Public License, where the Licensed
+     Material is a musical work, performance, or sound recording,
+     Adapted Material is always produced where the Licensed Material is
+     synched in timed relation with a moving image.
+
+  b. Adapter's License means the license You apply to Your Copyright
+     and Similar Rights in Your contributions to Adapted Material in
+     accordance with the terms and conditions of this Public License.
+
+  c. Copyright and Similar Rights means copyright and/or similar rights
+     closely related to copyright including, without limitation,
+     performance, broadcast, sound recording, and Sui Generis Database
+     Rights, without regard to how the rights are labeled or
+     categorized. For purposes of this Public License, the rights
+     specified in Section 2(b)(1)-(2) are not Copyright and Similar
+     Rights.
+
+  d. Effective Technological Measures means those measures that, in the
+     absence of proper authority, may not be circumvented under laws
+     fulfilling obligations under Article 11 of the WIPO Copyright
+     Treaty adopted on December 20, 1996, and/or similar international
+     agreements.
+
+  e. Exceptions and Limitations means fair use, fair dealing, and/or
+     any other exception or limitation to Copyright and Similar Rights
+     that applies to Your use of the Licensed Material.
+
+  f. Licensed Material means the artistic or literary work, database,
+     or other material to which the Licensor applied this Public
+     License.
+
+  g. Licensed Rights means the rights granted to You subject to the
+     terms and conditions of this Public License, which are limited to
+     all Copyright and Similar Rights that apply to Your use of the
+     Licensed Material and that the Licensor has authority to license.
+
+  h. Licensor means the individual(s) or entity(ies) granting rights
+     under this Public License.
+
+  i. Share means to provide material to the public by any means or
+     process that requires permission under the Licensed Rights, such
+     as reproduction, public display, public performance, distribution,
+     dissemination, communication, or importation, and to make material
+     available to the public including in ways that members of the
+     public may access the material from a place and at a time
+     individually chosen by them.
+
+  j. Sui Generis Database Rights means rights other than copyright
+     resulting from Directive 96/9/EC of the European Parliament and of
+     the Council of 11 March 1996 on the legal protection of databases,
+     as amended and/or succeeded, as well as other essentially
+     equivalent rights anywhere in the world.
+
+  k. You means the individual or entity exercising the Licensed Rights
+     under this Public License. Your has a corresponding meaning.
+
+
+Section 2 -- Scope.
+
+  a. License grant.
+
+       1. Subject to the terms and conditions of this Public License,
+          the Licensor hereby grants You a worldwide, royalty-free,
+          non-sublicensable, non-exclusive, irrevocable license to
+          exercise the Licensed Rights in the Licensed Material to:
+
+            a. reproduce and Share the Licensed Material, in whole or
+               in part; and
+
+            b. produce, reproduce, and Share Adapted Material.
+
+       2. Exceptions and Limitations. For the avoidance of doubt, where
+          Exceptions and Limitations apply to Your use, this Public
+          License does not apply, and You do not need to comply with
+          its terms and conditions.
+
+       3. Term. The term of this Public License is specified in Section
+          6(a).
+
+       4. Media and formats; technical modifications allowed. The
+          Licensor authorizes You to exercise the Licensed Rights in
+          all media and formats whether now known or hereafter created,
+          and to make technical modifications necessary to do so. The
+          Licensor waives and/or agrees not to assert any right or
+          authority to forbid You from making technical modifications
+          necessary to exercise the Licensed Rights, including
+          technical modifications necessary to circumvent Effective
+          Technological Measures. For purposes of this Public License,
+          simply making modifications authorized by this Section 2(a)
+          (4) never produces Adapted Material.
+
+       5. Downstream recipients.
+
+            a. Offer from the Licensor -- Licensed Material. Every
+               recipient of the Licensed Material automatically
+               receives an offer from the Licensor to exercise the
+               Licensed Rights under the terms and conditions of this
+               Public License.
+
+            b. No downstream restrictions. You may not offer or impose
+               any additional or different terms or conditions on, or
+               apply any Effective Technological Measures to, the
+               Licensed Material if doing so restricts exercise of the
+               Licensed Rights by any recipient of the Licensed
+               Material.
+
+       6. No endorsement. Nothing in this Public License constitutes or
+          may be construed as permission to assert or imply that You
+          are, or that Your use of the Licensed Material is, connected
+          with, or sponsored, endorsed, or granted official status by,
+          the Licensor or others designated to receive attribution as
+          provided in Section 3(a)(1)(A)(i).
+
+  b. Other rights.
+
+       1. Moral rights, such as the right of integrity, are not
+          licensed under this Public License, nor are publicity,
+          privacy, and/or other similar personality rights; however, to
+          the extent possible, the Licensor waives and/or agrees not to
+          assert any such rights held by the Licensor to the limited
+          extent necessary to allow You to exercise the Licensed
+          Rights, but not otherwise.
+
+       2. Patent and trademark rights are not licensed under this
+          Public License.
+
+       3. To the extent possible, the Licensor waives any right to
+          collect royalties from You for the exercise of the Licensed
+          Rights, whether directly or through a collecting society
+          under any voluntary or waivable statutory or compulsory
+          licensing scheme. In all other cases the Licensor expressly
+          reserves any right to collect such royalties.
+
+
+Section 3 -- License Conditions.
+
+Your exercise of the Licensed Rights is expressly made subject to the
+following conditions.
+
+  a. Attribution.
+
+       1. If You Share the Licensed Material (including in modified
+          form), You must:
+
+            a. retain the following if it is supplied by the Licensor
+               with the Licensed Material:
+
+                 i. identification of the creator(s) of the Licensed
+                    Material and any others designated to receive
+                    attribution, in any reasonable manner requested by
+                    the Licensor (including by pseudonym if
+                    designated);
+
+                ii. a copyright notice;
+
+               iii. a notice that refers to this Public License;
+
+                iv. a notice that refers to the disclaimer of
+                    warranties;
+
+                 v. a URI or hyperlink to the Licensed Material to the
+                    extent reasonably practicable;
+
+            b. indicate if You modified the Licensed Material and
+               retain an indication of any previous modifications; and
+
+            c. indicate the Licensed Material is licensed under this
+               Public License, and include the text of, or the URI or
+               hyperlink to, this Public License.
+
+       2. You may satisfy the conditions in Section 3(a)(1) in any
+          reasonable manner based on the medium, means, and context in
+          which You Share the Licensed Material. For example, it may be
+          reasonable to satisfy the conditions by providing a URI or
+          hyperlink to a resource that includes the required
+          information.
+
+       3. If requested by the Licensor, You must remove any of the
+          information required by Section 3(a)(1)(A) to the extent
+          reasonably practicable.
+
+       4. If You Share Adapted Material You produce, the Adapter's
+          License You apply must not prevent recipients of the Adapted
+          Material from complying with this Public License.
+
+
+Section 4 -- Sui Generis Database Rights.
+
+Where the Licensed Rights include Sui Generis Database Rights that
+apply to Your use of the Licensed Material:
+
+  a. for the avoidance of doubt, Section 2(a)(1) grants You the right
+     to extract, reuse, reproduce, and Share all or a substantial
+     portion of the contents of the database;
+
+  b. if You include all or a substantial portion of the database
+     contents in a database in which You have Sui Generis Database
+     Rights, then the database in which You have Sui Generis Database
+     Rights (but not its individual contents) is Adapted Material; and
+
+  c. You must comply with the conditions in Section 3(a) if You Share
+     all or a substantial portion of the contents of the database.
+
+For the avoidance of doubt, this Section 4 supplements and does not
+replace Your obligations under this Public License where the Licensed
+Rights include other Copyright and Similar Rights.
+
+
+Section 5 -- Disclaimer of Warranties and Limitation of Liability.
+
+  a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
+     EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
+     AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
+     ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
+     IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
+     WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
+     PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
+     ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
+     KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
+     ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
+
+  b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
+     TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
+     NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
+     INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
+     COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
+     USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
+     ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
+     DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
+     IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
+
+  c. The disclaimer of warranties and limitation of liability provided
+     above shall be interpreted in a manner that, to the extent
+     possible, most closely approximates an absolute disclaimer and
+     waiver of all liability.
+
+
+Section 6 -- Term and Termination.
+
+  a. This Public License applies for the term of the Copyright and
+     Similar Rights licensed here. However, if You fail to comply with
+     this Public License, then Your rights under this Public License
+     terminate automatically.
+
+  b. Where Your right to use the Licensed Material has terminated under
+     Section 6(a), it reinstates:
+
+       1. automatically as of the date the violation is cured, provided
+          it is cured within 30 days of Your discovery of the
+          violation; or
+
+       2. upon express reinstatement by the Licensor.
+
+     For the avoidance of doubt, this Section 6(b) does not affect any
+     right the Licensor may have to seek remedies for Your violations
+     of this Public License.
+
+  c. For the avoidance of doubt, the Licensor may also offer the
+     Licensed Material under separate terms or conditions or stop
+     distributing the Licensed Material at any time; however, doing so
+     will not terminate this Public License.
+
+  d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
+     License.
+
+
+Section 7 -- Other Terms and Conditions.
+
+  a. The Licensor shall not be bound by any additional or different
+     terms or conditions communicated by You unless expressly agreed.
+
+  b. Any arrangements, understandings, or agreements regarding the
+     Licensed Material not stated herein are separate from and
+     independent of the terms and conditions of this Public License.
+
+
+Section 8 -- Interpretation.
+
+  a. For the avoidance of doubt, this Public License does not, and
+     shall not be interpreted to, reduce, limit, restrict, or impose
+     conditions on any use of the Licensed Material that could lawfully
+     be made without permission under this Public License.
+
+  b. To the extent possible, if any provision of this Public License is
+     deemed unenforceable, it shall be automatically reformed to the
+     minimum extent necessary to make it enforceable. If the provision
+     cannot be reformed, it shall be severed from this Public License
+     without affecting the enforceability of the remaining terms and
+     conditions.
+
+  c. No term or condition of this Public License will be waived and no
+     failure to comply consented to unless expressly agreed to by the
+     Licensor.
+
+  d. Nothing in this Public License constitutes or may be interpreted
+     as a limitation upon, or waiver of, any privileges and immunities
+     that apply to the Licensor or You, including from the legal
+     processes of any jurisdiction or authority.
+
+
+=======================================================================
+
+Creative Commons is not a party to its public licenses.
+Notwithstanding, Creative Commons may elect to apply one of its public
+licenses to material it publishes and in those instances will be
+considered the "Licensor." Except for the limited purpose of indicating
+that material is shared under a Creative Commons public license or as
+otherwise permitted by the Creative Commons policies published at
+creativecommons.org/policies, Creative Commons does not authorize the
+use of the trademark "Creative Commons" or any other trademark or logo
+of Creative Commons without its prior written consent including,
+without limitation, in connection with any unauthorized modifications
+to any of its public licenses or any other arrangements,
+understandings, or agreements concerning use of licensed material. For
+the avoidance of doubt, this paragraph does not form part of the public
+licenses.
+
+Creative Commons may be contacted at creativecommons.org.

src/ndl_layout/mmdetection/README.md

@@ -0,0 +1,191 @@
+This software was developed by the National Diet Library under contract to Morpho AI Solutions, Inc.
+This software is largely based on the following repositories.
+
+- [open-mmlab/mmdetection](https://github.com/open-mmlab/mmdetection)
+
+The following files are also based on [eqlv2](https://github.com/tztztztztz/eqlv2)
+
+- [mmdet/core/post_processing/bbox_nms.py](mmdet/core/post_processing/bbox_nms.py)
+- [mmdet/core/post_processing/merge_augs.py](mmdet/core/post_processing/merge_augs.py)
+- [mmdet/datasets/builder.py](mmdet/datasets/builder.py)
+- [mmdet/datasets/class_balance_dataset_wrapper.py](mmdet/datasets/class_balance_dataset_wrapper.py)
+- [mmdet/datasets/max_iter_dataset_wrapper.py](mmdet/datasets/max_iter_dataset_wrapper.py)
+- [mmdet/models/losses/eql.py](mmdet/models/losses/eql.py)
+- [mmdet/models/losses/eqlv2.py](mmdet/models/losses/eqlv2.py)
+- [mmdet/models/losses/group_softmax.py](mmdet/models/losses/group_softmax.py)
+
+The newly developed portion of this program is released by the National Diet Library under a CC BY 4.0 license. For more information, see [LICENSE](./LICENSE)
+.
+
+<div align="center">
+  <img src="resources/mmdet-logo.png" width="600"/>
+</div>
+
+**News**: We released the technical report on [ArXiv](https://arxiv.org/abs/1906.07155).
+
+Documentation: https://mmdetection.readthedocs.io/
+
+## Introduction
+
+English | [简体中文](README_zh-CN.md)
+
+MMDetection is an open source object detection toolbox based on PyTorch. It is
+a part of the [OpenMMLab](https://openmmlab.com/) project.
+
+The master branch works with **PyTorch 1.3+**.
+The old v1.x branch works with PyTorch 1.1 to 1.4, but v2.0 is strongly recommended for faster speed, higher performance, better design and more friendly usage.
+
+![demo image](resources/coco_test_12510.jpg)
+
+### Major features
+
+- **Modular Design**
+
+  We decompose the detection framework into different components and one can easily construct a customized object detection framework by combining different modules.
+
+- **Support of multiple frameworks out of box**
+
+  The toolbox directly supports popular and contemporary detection frameworks, *e.g.* Faster RCNN, Mask RCNN, RetinaNet, etc.
+
+- **High efficiency**
+
+  All basic bbox and mask operations run on GPUs. The training speed is faster than or comparable to other codebases, including [Detectron2](https://github.com/facebookresearch/detectron2), [maskrcnn-benchmark](https://github.com/facebookresearch/maskrcnn-benchmark) and [SimpleDet](https://github.com/TuSimple/simpledet).
+
+- **State of the art**
+
+  The toolbox stems from the codebase developed by the *MMDet* team, who won [COCO Detection Challenge](http://cocodataset.org/#detection-leaderboard) in 2018, and we keep pushing it forward.
+
+Apart from MMDetection, we also released a library [mmcv](https://github.com/open-mmlab/mmcv) for computer vision research, which is heavily depended on by this toolbox.
+
+## License
+
+This project is released under the [Apache 2.0 license](LICENSE).
+
+## Changelog
+
+v2.11.0 was released in 01/04/2021.
+Please refer to [changelog.md](docs/changelog.md) for details and release history.
+A comparison between v1.x and v2.0 codebases can be found in [compatibility.md](docs/compatibility.md).
+
+## Benchmark and model zoo
+
+Results and models are available in the [model zoo](docs/model_zoo.md).
+
+Supported backbones:
+
+- [x] ResNet (CVPR'2016)
+- [x] ResNeXt (CVPR'2017)
+- [x] VGG (ICLR'2015)
+- [x] HRNet (CVPR'2019)
+- [x] RegNet (CVPR'2020)
+- [x] Res2Net (TPAMI'2020)
+- [x] ResNeSt (ArXiv'2020)
+
+Supported methods:
+
+- [x] [RPN (NeurIPS'2015)](configs/rpn)
+- [x] [Fast R-CNN (ICCV'2015)](configs/fast_rcnn)
+- [x] [Faster R-CNN (NeurIPS'2015)](configs/faster_rcnn)
+- [x] [Mask R-CNN (ICCV'2017)](configs/mask_rcnn)
+- [x] [Cascade R-CNN (CVPR'2018)](configs/cascade_rcnn)
+- [x] [Cascade Mask R-CNN (CVPR'2018)](configs/cascade_rcnn)
+- [x] [SSD (ECCV'2016)](configs/ssd)
+- [x] [RetinaNet (ICCV'2017)](configs/retinanet)
+- [x] [GHM (AAAI'2019)](configs/ghm)
+- [x] [Mask Scoring R-CNN (CVPR'2019)](configs/ms_rcnn)
+- [x] [Double-Head R-CNN (CVPR'2020)](configs/double_heads)
+- [x] [Hybrid Task Cascade (CVPR'2019)](configs/htc)
+- [x] [Libra R-CNN (CVPR'2019)](configs/libra_rcnn)
+- [x] [Guided Anchoring (CVPR'2019)](configs/guided_anchoring)
+- [x] [FCOS (ICCV'2019)](configs/fcos)
+- [x] [RepPoints (ICCV'2019)](configs/reppoints)
+- [x] [Foveabox (TIP'2020)](configs/foveabox)
+- [x] [FreeAnchor (NeurIPS'2019)](configs/free_anchor)
+- [x] [NAS-FPN (CVPR'2019)](configs/nas_fpn)
+- [x] [ATSS (CVPR'2020)](configs/atss)
+- [x] [FSAF (CVPR'2019)](configs/fsaf)
+- [x] [PAFPN (CVPR'2018)](configs/pafpn)
+- [x] [Dynamic R-CNN (ECCV'2020)](configs/dynamic_rcnn)
+- [x] [PointRend (CVPR'2020)](configs/point_rend)
+- [x] [CARAFE (ICCV'2019)](configs/carafe/README.md)
+- [x] [DCNv2 (CVPR'2019)](configs/dcn/README.md)
+- [x] [Group Normalization (ECCV'2018)](configs/gn/README.md)
+- [x] [Weight Standardization (ArXiv'2019)](configs/gn+ws/README.md)
+- [x] [OHEM (CVPR'2016)](configs/faster_rcnn/faster_rcnn_r50_fpn_ohem_1x_coco.py)
+- [x] [Soft-NMS (ICCV'2017)](configs/faster_rcnn/faster_rcnn_r50_fpn_soft_nms_1x_coco.py)
+- [x] [Generalized Attention (ICCV'2019)](configs/empirical_attention/README.md)
+- [x] [GCNet (ICCVW'2019)](configs/gcnet/README.md)
+- [x] [Mixed Precision (FP16) Training (ArXiv'2017)](configs/fp16/README.md)
+- [x] [InstaBoost (ICCV'2019)](configs/instaboost/README.md)
+- [x] [GRoIE (ICPR'2020)](configs/groie/README.md)
+- [x] [DetectoRS (ArXix'2020)](configs/detectors/README.md)
+- [x] [Generalized Focal Loss (NeurIPS'2020)](configs/gfl/README.md)
+- [x] [CornerNet (ECCV'2018)](configs/cornernet/README.md)
+- [x] [Side-Aware Boundary Localization (ECCV'2020)](configs/sabl/README.md)
+- [x] [YOLOv3 (ArXiv'2018)](configs/yolo/README.md)
+- [x] [PAA (ECCV'2020)](configs/paa/README.md)
+- [x] [YOLACT (ICCV'2019)](configs/yolact/README.md)
+- [x] [CentripetalNet (CVPR'2020)](configs/centripetalnet/README.md)
+- [x] [VFNet (ArXix'2020)](configs/vfnet/README.md)
+- [x] [DETR (ECCV'2020)](configs/detr/README.md)
+- [x] [Deformable DETR (ICLR'2021)](configs/deformable_detr/README.md)
+- [x] [CascadeRPN (NeurIPS'2019)](configs/cascade_rpn/README.md)
+- [x] [SCNet (AAAI'2021)](configs/scnet/README.md)
+- [x] [AutoAssign (ArXix'2020)](configs/autoassign/README.md)
+- [x] [YOLOF (CVPR'2021)](configs/yolof/README.md)
+
+
+Some other methods are also supported in [projects using MMDetection](./docs/projects.md).
+
+## Installation
+
+Please refer to [get_started.md](docs/get_started.md) for installation.
+
+## Getting Started
+
+Please see [get_started.md](docs/get_started.md) for the basic usage of MMDetection.
+We provide [colab tutorial](demo/MMDet_Tutorial.ipynb), and full guidance for quick run [with existing dataset](docs/1_exist_data_model.md) and [with new dataset](docs/2_new_data_model.md) for beginners.
+There are also tutorials for [finetuning models](docs/tutorials/finetune.md), [adding new dataset](docs/tutorials/new_dataset.md), [designing data pipeline](docs/tutorials/data_pipeline.md), [customizing models](docs/tutorials/customize_models.md), [customizing runtime settings](docs/tutorials/customize_runtime.md) and [useful tools](docs/useful_tools.md).
+
+Please refer to [FAQ](docs/faq.md) for frequently asked questions.
+
+## Contributing
+
+We appreciate all contributions to improve MMDetection. Please refer to [CONTRIBUTING.md](.github/CONTRIBUTING.md) for the contributing guideline.
+
+## Acknowledgement
+
+MMDetection is an open source project that is contributed by researchers and engineers from various colleges and companies. We appreciate all the contributors who implement their methods or add new features, as well as users who give valuable feedbacks.
+We wish that the toolbox and benchmark could serve the growing research community by providing a flexible toolkit to reimplement existing methods and develop their own new detectors.
+
+## Citation
+
+If you use this toolbox or benchmark in your research, please cite this project.
+
+```
+@article{mmdetection,
+  title   = {{MMDetection}: Open MMLab Detection Toolbox and Benchmark},
+  author  = {Chen, Kai and Wang, Jiaqi and Pang, Jiangmiao and Cao, Yuhang and
+             Xiong, Yu and Li, Xiaoxiao and Sun, Shuyang and Feng, Wansen and
+             Liu, Ziwei and Xu, Jiarui and Zhang, Zheng and Cheng, Dazhi and
+             Zhu, Chenchen and Cheng, Tianheng and Zhao, Qijie and Li, Buyu and
+             Lu, Xin and Zhu, Rui and Wu, Yue and Dai, Jifeng and Wang, Jingdong
+             and Shi, Jianping and Ouyang, Wanli and Loy, Chen Change and Lin, Dahua},
+  journal= {arXiv preprint arXiv:1906.07155},
+  year={2019}
+}
+```
+
+## Projects in OpenMMLab
+
+- [MMCV](https://github.com/open-mmlab/mmcv): OpenMMLab foundational library for computer vision.
+- [MMClassification](https://github.com/open-mmlab/mmclassification): OpenMMLab image classification toolbox and benchmark.
+- [MMDetection](https://github.com/open-mmlab/mmdetection): OpenMMLab detection toolbox and benchmark.
+- [MMDetection3D](https://github.com/open-mmlab/mmdetection3d): OpenMMLab's next-generation platform for general 3D object detection.
+- [MMSegmentation](https://github.com/open-mmlab/mmsegmentation): OpenMMLab semantic segmentation toolbox and benchmark.
+- [MMAction2](https://github.com/open-mmlab/mmaction2): OpenMMLab's next-generation action understanding toolbox and benchmark.
+- [MMTracking](https://github.com/open-mmlab/mmtracking): OpenMMLab video perception toolbox and benchmark.
+- [MMPose](https://github.com/open-mmlab/mmpose): OpenMMLab pose estimation toolbox and benchmark.
+- [MMEditing](https://github.com/open-mmlab/mmediting): OpenMMLab image and video editing toolbox.
+- [MMOCR](https://github.com/open-mmlab/mmocr): A Comprehensive Toolbox for Text Detection, Recognition and Understanding.
+- [MMGeneration](https://github.com/open-mmlab/mmgeneration): OpenMMLab image and video generative models toolbox.

src/ndl_layout/mmdetection/README_zh-CN.md

@@ -0,0 +1,190 @@
+<div align="center">
+  <img src="resources/mmdet-logo.png" width="600"/>
+</div>
+
+**新闻**: 我们在 [ArXiv](https://arxiv.org/abs/1906.07155) 上公开了技术报告。
+
+文档: https://mmdetection.readthedocs.io/
+
+## 简介
+
+[English](README.md) | 简体中文
+
+MMDetection 是一个基于 PyTorch 的目标检测开源工具箱。它是 [OpenMMLab](https://openmmlab.com/) 项目的一部分。
+
+主分支代码目前支持 PyTorch 1.3 以上的版本。
+
+v1.x 的历史版本支持 PyTorch 1.1 到 1.4，但是我们强烈建议用户使用新的 2.x 的版本，新的版本速度更快，性能更高，有更优雅的代码设计，对用户使用也更加友好。
+
+![demo image](resources/coco_test_12510.jpg)
+
+### 主要特性
+
+- **模块化设计**
+
+  MMDetection 将检测框架解耦成不同的模块组件，通过组合不同的模块组件，用户可以便捷地构建自定义的检测模型
+
+- **丰富的即插即用的算法和模型**
+
+  MMDetection 支持了众多主流的和最新的检测算法，例如 Faster R-CNN，Mask R-CNN，RetinaNet 等。
+
+- **速度快**
+
+  基本的框和 mask 操作都实现了 GPU 版本，训练速度比其他代码库更快或者相当，包括 [Detectron2](https://github.com/facebookresearch/detectron2), [maskrcnn-benchmark](https://github.com/facebookresearch/maskrcnn-benchmark) 和 [SimpleDet](https://github.com/TuSimple/simpledet)。
+
+- **性能高**
+
+  MMDetection 这个算法库源自于 COCO 2018 目标检测竞赛的冠军团队 *MMDet* 团队开发的代码，我们在之后持续进行了改进和提升。
+
+除了 MMDetection 之外，我们还开源了计算机视觉基础库 [MMCV](https://github.com/open-mmlab/mmcv)，MMCV 是 MMDetection 的主要依赖。
+
+## 开源许可证
+
+该项目采用 [Apache 2.0 开源许可证](LICENSE)。
+
+## 更新日志
+
+最新的月度版本 v2.11.0 在 2021.04.01 发布。
+如果想了解更多版本更新细节和历史信息，请阅读[更新日志](docs/changelog.md)。
+在[兼容性说明文档](docs/compatibility.md)中我们提供了 1.x 和 2.0 版本的详细比较。
+
+## 基准测试和模型库
+
+测试结果和模型可以在[模型库](docs/model_zoo.md)中找到。
+
+已支持的骨干网络：
+
+- [x] ResNet (CVPR'2016)
+- [x] ResNeXt (CVPR'2017)
+- [x] VGG (ICLR'2015)
+- [x] HRNet (CVPR'2019)
+- [x] RegNet (CVPR'2020)
+- [x] Res2Net (TPAMI'2020)
+- [x] ResNeSt (ArXiv'2020)
+
+已支持的算法：
+
+- [x] [RPN (NeurIPS'2015)](configs/rpn)
+- [x] [Fast R-CNN (ICCV'2015)](configs/fast_rcnn)
+- [x] [Faster R-CNN (NeurIPS'2015)](configs/faster_rcnn)
+- [x] [Mask R-CNN (ICCV'2017)](configs/mask_rcnn)
+- [x] [Cascade R-CNN (CVPR'2018)](configs/cascade_rcnn)
+- [x] [Cascade Mask R-CNN (CVPR'2018)](configs/cascade_rcnn)
+- [x] [SSD (ECCV'2016)](configs/ssd)
+- [x] [RetinaNet (ICCV'2017)](configs/retinanet)
+- [x] [GHM (AAAI'2019)](configs/ghm)
+- [x] [Mask Scoring R-CNN (CVPR'2019)](configs/ms_rcnn)
+- [x] [Double-Head R-CNN (CVPR'2020)](configs/double_heads)
+- [x] [Hybrid Task Cascade (CVPR'2019)](configs/htc)
+- [x] [Libra R-CNN (CVPR'2019)](configs/libra_rcnn)
+- [x] [Guided Anchoring (CVPR'2019)](configs/guided_anchoring)
+- [x] [FCOS (ICCV'2019)](configs/fcos)
+- [x] [RepPoints (ICCV'2019)](configs/reppoints)
+- [x] [Foveabox (TIP'2020)](configs/foveabox)
+- [x] [FreeAnchor (NeurIPS'2019)](configs/free_anchor)
+- [x] [NAS-FPN (CVPR'2019)](configs/nas_fpn)
+- [x] [ATSS (CVPR'2020)](configs/atss)
+- [x] [FSAF (CVPR'2019)](configs/fsaf)
+- [x] [PAFPN (CVPR'2018)](configs/pafpn)
+- [x] [Dynamic R-CNN (ECCV'2020)](configs/dynamic_rcnn)
+- [x] [PointRend (CVPR'2020)](configs/point_rend)
+- [x] [CARAFE (ICCV'2019)](configs/carafe/README.md)
+- [x] [DCNv2 (CVPR'2019)](configs/dcn/README.md)
+- [x] [Group Normalization (ECCV'2018)](configs/gn/README.md)
+- [x] [Weight Standardization (ArXiv'2019)](configs/gn+ws/README.md)
+- [x] [OHEM (CVPR'2016)](configs/faster_rcnn/faster_rcnn_r50_fpn_ohem_1x_coco.py)
+- [x] [Soft-NMS (ICCV'2017)](configs/faster_rcnn/faster_rcnn_r50_fpn_soft_nms_1x_coco.py)
+- [x] [Generalized Attention (ICCV'2019)](configs/empirical_attention/README.md)
+- [x] [GCNet (ICCVW'2019)](configs/gcnet/README.md)
+- [x] [Mixed Precision (FP16) Training (ArXiv'2017)](configs/fp16/README.md)
+- [x] [InstaBoost (ICCV'2019)](configs/instaboost/README.md)
+- [x] [GRoIE (ICPR'2020)](configs/groie/README.md)
+- [x] [DetectoRS (ArXix'2020)](configs/detectors/README.md)
+- [x] [Generalized Focal Loss (NeurIPS'2020)](configs/gfl/README.md)
+- [x] [CornerNet (ECCV'2018)](configs/cornernet/README.md)
+- [x] [Side-Aware Boundary Localization (ECCV'2020)](configs/sabl/README.md)
+- [x] [YOLOv3 (ArXiv'2018)](configs/yolo/README.md)
+- [x] [PAA (ECCV'2020)](configs/paa/README.md)
+- [x] [YOLACT (ICCV'2019)](configs/yolact/README.md)
+- [x] [CentripetalNet (CVPR'2020)](configs/centripetalnet/README.md)
+- [x] [VFNet (ArXix'2020)](configs/vfnet/README.md)
+- [x] [DETR (ECCV'2020)](configs/detr/README.md)
+- [x] [Deformable DETR (ICLR'2021)](configs/deformable_detr/README.md)
+- [x] [CascadeRPN (NeurIPS'2019)](configs/cascade_rpn/README.md)
+- [x] [SCNet (AAAI'2021)](configs/scnet/README.md)
+- [x] [AutoAssign (ArXix'2020)](configs/autoassign/README.md)
+- [x] [YOLOF (CVPR'2021)](configs/yolof/README.md)
+
+我们在[基于 MMDetection 的项目](./docs/projects.md)中列举了一些其他的支持的算法。
+
+## 安装
+
+请参考[快速入门文档](docs/get_started.md)进行安装。
+
+## 快速入门
+
+请参考[快速入门文档](docs/get_started.md)学习 MMDetection 的基本使用。
+我们提供了 [colab 教程](demo/MMDet_Tutorial.ipynb)，也为新手提供了完整的运行教程，分别针对[已有数据集](docs/1_exist_data_model.md)和[新数据集](docs/2_new_data_model.md) 完整的使用指南
+
+我们也提供了一些进阶教程，内容覆盖了 [finetune 模型](docs/tutorials/finetune.md)，[增加新数据集支持](docs/tutorials/new_dataset.md)，[设计新的数据预处理流程](docs/tutorials/data_pipeline.md)，[增加自定义模型](ocs/tutorials/customize_models.md)，[增加自定义的运行时配置](docs/tutorials/customize_runtime.md)，[常用工具和脚本](docs/useful_tools.md)。
+
+如果遇到问题，请参考 [FAQ 页面](docs/faq.md)。
+
+## 贡献指南
+
+我们感谢所有的贡献者为改进和提升 MMDetection 所作出的努力。请参考[贡献指南](.github/CONTRIBUTING.md)来了解参与项目贡献的相关指引。
+
+## 致谢
+
+MMDetection 是一款由来自不同高校和企业的研发人员共同参与贡献的开源项目。我们感谢所有为项目提供算法复现和新功能支持的贡献者，以及提供宝贵反馈的用户。 我们希望这个工具箱和基准测试可以为社区提供灵活的代码工具，供用户复现已有算法并开发自己的新模型，从而不断为开源社区提供贡献。
+
+## 引用
+
+如果你在研究中使用了本项目的代码或者性能基准，请参考如下 bibtex 引用 MMDetection。
+
+```
+@article{mmdetection,
+  title   = {{MMDetection}: Open MMLab Detection Toolbox and Benchmark},
+  author  = {Chen, Kai and Wang, Jiaqi and Pang, Jiangmiao and Cao, Yuhang and
+             Xiong, Yu and Li, Xiaoxiao and Sun, Shuyang and Feng, Wansen and
+             Liu, Ziwei and Xu, Jiarui and Zhang, Zheng and Cheng, Dazhi and
+             Zhu, Chenchen and Cheng, Tianheng and Zhao, Qijie and Li, Buyu and
+             Lu, Xin and Zhu, Rui and Wu, Yue and Dai, Jifeng and Wang, Jingdong
+             and Shi, Jianping and Ouyang, Wanli and Loy, Chen Change and Lin, Dahua},
+  journal= {arXiv preprint arXiv:1906.07155},
+  year={2019}
+}
+```
+
+## OpenMMLab 的其他项目
+
+- [MMCV](https://github.com/open-mmlab/mmcv): OpenMMLab 计算机视觉基础库
+- [MMClassification](https://github.com/open-mmlab/mmclassification): OpenMMLab 图像分类工具箱
+- [MMDetection](https://github.com/open-mmlab/mmdetection): OpenMMLab 目标检测工具箱
+- [MMDetection3D](https://github.com/open-mmlab/mmdetection3d): OpenMMLab 新一代通用 3D 目标检测平台
+- [MMSegmentation](https://github.com/open-mmlab/mmsegmentation): OpenMMLab 语义分割工具箱
+- [MMAction2](https://github.com/open-mmlab/mmaction2): OpenMMLab 新一代视频理解工具箱
+- [MMTracking](https://github.com/open-mmlab/mmtracking): OpenMMLab 一体化视频目标感知平台
+- [MMPose](https://github.com/open-mmlab/mmpose): OpenMMLab 姿态估计工具箱
+- [MMEditing](https://github.com/open-mmlab/mmediting): OpenMMLab 图像视频编辑工具箱
+- [MMOCR](https://github.com/open-mmlab/mmocr): OpenMMLab 全流程文字检测识别理解工具包
+- [MMGeneration](https://github.com/open-mmlab/mmgeneration): OpenMMLab 图片视频生成模型工具箱
+
+## 欢迎加入 OpenMMLab 社区
+
+扫描下方的二维码可关注 OpenMMLab 团队的 [知乎官方账号](https://www.zhihu.com/people/openmmlab)，加入 OpenMMLab 团队的 [官方交流 QQ 群](https://jq.qq.com/?_wv=1027&k=aCvMxdr3)
+
+<div align="center">
+<img src="/resources/zhihu_qrcode.jpg" height="400" />  <img src="/resources/qq_group_qrcode.jpg" height="400" />
+</div>
+
+我们会在 OpenMMLab 社区为大家
+
+- 📢 分享 AI 框架的前沿核心技术
+- 💻 解读 PyTorch 常用模块源码
+- 📰 发布 OpenMMLab 的相关新闻
+- 🚀 介绍 OpenMMLab 开发的前沿算法
+- 🏃 获取更高效的问题答疑和意见反馈
+- 🔥 提供与各行各业开发者充分交流的平台
+
+干货满满 📘，等你来撩 💗，OpenMMLab 社区期待您的加入 👬

src/ndl_layout/mmdetection/configs/_base_/datasets/cityscapes_detection.py

@@ -0,0 +1,56 @@
+# dataset settings
+dataset_type = 'CityscapesDataset'
+data_root = 'data/cityscapes/'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations', with_bbox=True),
+    dict(
+        type='Resize', img_scale=[(2048, 800), (2048, 1024)], keep_ratio=True),
+    dict(type='RandomFlip', flip_ratio=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='Pad', size_divisor=32),
+    dict(type='DefaultFormatBundle'),
+    dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels']),
+]
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(
+        type='MultiScaleFlipAug',
+        img_scale=(2048, 1024),
+        flip=False,
+        transforms=[
+            dict(type='Resize', keep_ratio=True),
+            dict(type='RandomFlip'),
+            dict(type='Normalize', **img_norm_cfg),
+            dict(type='Pad', size_divisor=32),
+            dict(type='ImageToTensor', keys=['img']),
+            dict(type='Collect', keys=['img']),
+        ])
+]
+data = dict(
+    samples_per_gpu=1,
+    workers_per_gpu=2,
+    train=dict(
+        type='RepeatDataset',
+        times=8,
+        dataset=dict(
+            type=dataset_type,
+            ann_file=data_root +
+            'annotations/instancesonly_filtered_gtFine_train.json',
+            img_prefix=data_root + 'leftImg8bit/train/',
+            pipeline=train_pipeline)),
+    val=dict(
+        type=dataset_type,
+        ann_file=data_root +
+        'annotations/instancesonly_filtered_gtFine_val.json',
+        img_prefix=data_root + 'leftImg8bit/val/',
+        pipeline=test_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=data_root +
+        'annotations/instancesonly_filtered_gtFine_test.json',
+        img_prefix=data_root + 'leftImg8bit/test/',
+        pipeline=test_pipeline))
+evaluation = dict(interval=1, metric='bbox')

src/ndl_layout/mmdetection/configs/_base_/datasets/cityscapes_instance.py

@@ -0,0 +1,56 @@
+# dataset settings
+dataset_type = 'CityscapesDataset'
+data_root = 'data/cityscapes/'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
+    dict(
+        type='Resize', img_scale=[(2048, 800), (2048, 1024)], keep_ratio=True),
+    dict(type='RandomFlip', flip_ratio=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='Pad', size_divisor=32),
+    dict(type='DefaultFormatBundle'),
+    dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks']),
+]
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(
+        type='MultiScaleFlipAug',
+        img_scale=(2048, 1024),
+        flip=False,
+        transforms=[
+            dict(type='Resize', keep_ratio=True),
+            dict(type='RandomFlip'),
+            dict(type='Normalize', **img_norm_cfg),
+            dict(type='Pad', size_divisor=32),
+            dict(type='ImageToTensor', keys=['img']),
+            dict(type='Collect', keys=['img']),
+        ])
+]
+data = dict(
+    samples_per_gpu=1,
+    workers_per_gpu=2,
+    train=dict(
+        type='RepeatDataset',
+        times=8,
+        dataset=dict(
+            type=dataset_type,
+            ann_file=data_root +
+            'annotations/instancesonly_filtered_gtFine_train.json',
+            img_prefix=data_root + 'leftImg8bit/train/',
+            pipeline=train_pipeline)),
+    val=dict(
+        type=dataset_type,
+        ann_file=data_root +
+        'annotations/instancesonly_filtered_gtFine_val.json',
+        img_prefix=data_root + 'leftImg8bit/val/',
+        pipeline=test_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=data_root +
+        'annotations/instancesonly_filtered_gtFine_test.json',
+        img_prefix=data_root + 'leftImg8bit/test/',
+        pipeline=test_pipeline))
+evaluation = dict(metric=['bbox', 'segm'])

src/ndl_layout/mmdetection/configs/_base_/datasets/coco_detection.py

@@ -0,0 +1,49 @@
+# dataset settings
+dataset_type = 'CocoDataset'
+data_root = 'data/coco/'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations', with_bbox=True),
+    dict(type='Resize', img_scale=(1333, 800), keep_ratio=True),
+    dict(type='RandomFlip', flip_ratio=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='Pad', size_divisor=32),
+    dict(type='DefaultFormatBundle'),
+    dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels']),
+]
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(
+        type='MultiScaleFlipAug',
+        img_scale=(1333, 800),
+        flip=False,
+        transforms=[
+            dict(type='Resize', keep_ratio=True),
+            dict(type='RandomFlip'),
+            dict(type='Normalize', **img_norm_cfg),
+            dict(type='Pad', size_divisor=32),
+            dict(type='ImageToTensor', keys=['img']),
+            dict(type='Collect', keys=['img']),
+        ])
+]
+data = dict(
+    samples_per_gpu=2,
+    workers_per_gpu=2,
+    train=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_train2017.json',
+        img_prefix=data_root + 'train2017/',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_val2017.json',
+        img_prefix=data_root + 'val2017/',
+        pipeline=test_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_val2017.json',
+        img_prefix=data_root + 'val2017/',
+        pipeline=test_pipeline))
+evaluation = dict(interval=1, metric='bbox')

src/ndl_layout/mmdetection/configs/_base_/datasets/coco_instance.py

@@ -0,0 +1,49 @@
+# dataset settings
+dataset_type = 'CocoDataset'
+data_root = 'data/coco/'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
+    dict(type='Resize', img_scale=(1333, 800), keep_ratio=True),
+    dict(type='RandomFlip', flip_ratio=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='Pad', size_divisor=32),
+    dict(type='DefaultFormatBundle'),
+    dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks']),
+]
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(
+        type='MultiScaleFlipAug',
+        img_scale=(1333, 800),
+        flip=False,
+        transforms=[
+            dict(type='Resize', keep_ratio=True),
+            dict(type='RandomFlip'),
+            dict(type='Normalize', **img_norm_cfg),
+            dict(type='Pad', size_divisor=32),
+            dict(type='ImageToTensor', keys=['img']),
+            dict(type='Collect', keys=['img']),
+        ])
+]
+data = dict(
+    samples_per_gpu=2,
+    workers_per_gpu=2,
+    train=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_train2017.json',
+        img_prefix=data_root + 'train2017/',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_val2017.json',
+        img_prefix=data_root + 'val2017/',
+        pipeline=test_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_val2017.json',
+        img_prefix=data_root + 'val2017/',
+        pipeline=test_pipeline))
+evaluation = dict(metric=['bbox', 'segm'])