Spaces:
Running
Running
| import streamlit as st | |
| import fitz # PyMuPDF | |
| from PIL import Image | |
| import pytesseract | |
| import numpy as np | |
| from streamlit_drawable_canvas import st_canvas | |
| import io | |
| def pdf_page_to_image(doc, page_number=0, scale=1.0): | |
| page = doc.load_page(page_number) | |
| pix = page.get_pixmap(matrix=fitz.Matrix(scale, scale)) | |
| img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples) | |
| gray_img = img.convert("L") | |
| return gray_img | |
| def extract_text_tesseract(image): | |
| """Use Tesseract to extract text from an image.""" | |
| return pytesseract.image_to_string(image) | |
| def main(): | |
| st.sidebar.title("PDF Navigation") | |
| pdf_file = st.sidebar.file_uploader("Upload a PDF file", type=["pdf"]) | |
| if pdf_file: | |
| doc = fitz.open("pdf", pdf_file.getvalue()) | |
| total_pages = doc.page_count | |
| selected_page = st.sidebar.slider("Select a Page", 1, total_pages, 1) - 1 | |
| zoom_factor = st.sidebar.slider("Zoom Factor", 0.5, 3.0, 1.0, 0.1) | |
| img = pdf_page_to_image(doc, page_number=selected_page, scale=zoom_factor) | |
| img_array = np.array(img) | |
| # Container to add scrollbars | |
| container = st.container() | |
| with container: | |
| st.image(img_array, use_column_width=True, caption=f"Page {selected_page + 1}") | |
| canvas_result = st_canvas( | |
| fill_color="rgba(255, 165, 0, 0.3)", | |
| stroke_width=0, | |
| stroke_color="#ffffff", | |
| background_image=Image.fromarray(img_array), | |
| update_streamlit=True, | |
| height=int(img.height), | |
| width=int(img.width), | |
| drawing_mode="rect", | |
| key="canvas" + str(selected_page) + str(zoom_factor), | |
| ) | |
| if st.button("Extract Text from Selected Region"): | |
| selected_areas = len(canvas_result.json_data["objects"]) | |
| texts = [] | |
| for area_id in range(selected_areas): | |
| bbox = canvas_result.json_data["objects"][area_id] if canvas_result.json_data["objects"] else None | |
| if bbox: | |
| x, y, w, h = bbox['left'], bbox['top'], bbox['width'], bbox['height'] | |
| rect = [int(x), int(y), int(x + w), int(y + h)] | |
| img_crop = img.crop(rect) | |
| text = extract_text_tesseract(img_crop) | |
| texts.append(text) | |
| for id, text in enumerate(texts): | |
| st.write(f"Extracted Text from selection {id}:") | |
| st.write(text) | |
| doc.close() | |
| if __name__ == "__main__": | |
| main() |