import streamlit as st import gradio as gr import cv2 import numpy as np from tensorflow.keras.models import load_model from tensorflow.keras.applications.vgg16 import preprocess_input from tensorflow.keras.preprocessing import image from werkzeug.utils import secure_filename import os # Loading Models braintumor_model = load_model('models/brain_tumor_binary.h5') # Configuring Streamlit UPLOAD_FOLDER = 'static/uploads' ALLOWED_EXTENSIONS = {'png', 'jpg', 'jpeg'} st.set_page_config(page_title="Brain Tumor Prediction App", page_icon=":brain:") def allowed_file(filename): return '.' in filename and filename.rsplit('.', 1)[1].lower() in ALLOWED_EXTENSIONS def preprocess_imgs(set_name, img_size): set_new = [] for img in set_name: img = cv2.resize(img, dsize=img_size, interpolation=cv2.INTER_CUBIC) set_new.append(preprocess_input(img)) return np.array(set_new) def crop_imgs(set_name, add_pixels_value=0): set_new = [] for img in set_name: gray = cv2.GaussianBlur(img, (5, 5), 0) thresh = cv2.threshold(gray, 45, 255, cv2.THRESH_BINARY)[1] thresh = cv2.erode(thresh, None, iterations=2) thresh = cv2.dilate(thresh, None, iterations=2) cnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE) cnts = cnts[0] if len(cnts) == 2 else cnts[1] c = max(cnts, key=cv2.contourArea) extLeft = tuple(c[c[:, :, 0].argmin()][0]) extRight = tuple(c[c[:, :, 0].argmax()][0]) extTop = tuple(c[c[:, :, 1].argmin()][0]) extBot = tuple(c[c[:, :, 1].argmax()][0]) ADD_PIXELS = add_pixels_value new_img = img[extTop[1] - ADD_PIXELS:extBot[1] + ADD_PIXELS, extLeft[0] - ADD_PIXELS:extRight[0] + ADD_PIXELS].copy() set_new.append(new_img) return np.array(set_new) # Function to preprocess the image def preprocess_image(img_array): img_array = cv2.cvtColor(img_array, cv2.COLOR_BGR2RGB) img = image.array_to_img(img_array, target_size=(200, 200)) img_array = image.img_to_array(img) img_array = np.expand_dims(img_array, axis=0) img_array /= 255.0 # Normalize the image return img_array def predict_braintumor(img_input): if isinstance(img_input, str): # File path (for Streamlit) img_path = img_input img_gray = cv2.imread(img_path, cv2.IMREAD_GRAYSCALE) else: # NumPy array (for Gradio) img_array = img_input.astype(np.uint8) img_gray = cv2.cvtColor(img_array, cv2.COLOR_BGR2GRAY) # Crop and preprocess the grayscale image img_processed = crop_imgs([img_gray]) img_processed = preprocess_imgs(img_processed, (224, 224)) # Make prediction pred = braintumor_model.predict(img_processed) # Handle binary decision pred = 1 if pred[0] >= 0.5 else 0 return pred def main(): st.title("Brain Tumor Prediction App") uploaded_file = st.file_uploader("Choose an image...", type=["png", "jpg", "jpeg"]) if uploaded_file is not None: st.image(uploaded_file, caption="Uploaded Image.", use_column_width=True) st.write("") st.write("Classifying...") # Read the contents of the uploaded file or the input image file_contents = uploaded_file.read() uploaded_array = np.asarray(bytearray(file_contents), dtype=np.uint8) uploaded_img = cv2.imdecode(uploaded_array, -1) # Save the uploaded file if isinstance(file_contents, bytes): filename = secure_filename(uploaded_file.name) file_path = os.path.join(UPLOAD_FOLDER, filename) with open(file_path, "wb") as f: f.write(file_contents) # Make prediction result = predict_braintumor(file_path) else: # Make prediction for Gradio (direct input of image) result = predict_braintumor(uploaded_img) # Handle binary decision result_text = "Brain Tumor Found!" if result == 1 else "Brain Tumor Not Found!" # Display prediction st.subheader("Prediction:") st.success(result_text) if __name__ == "__main__": # Streamlit part if "streamlit" in st.__version__.lower(): main() # Gradio part iface = gr.Interface( fn=predict_braintumor, inputs="image", outputs="text", examples=[["examples/1 no.jpg"]], enable_queue=True ) iface.launch()