import cv2 import numpy as np from pdf2image import pdf2image from vidocp.utils.detection import detect_large_coherent_structures from vidocp.utils.display import show_mpl from vidocp.utils.draw import draw_rectangles from vidocp.utils.post_processing import remove_included from vidocp.utils.filters import is_large_enough, has_acceptable_format from vidocp.utils.text import remove_primary_text_regions def is_likely_figure(cont, min_area=5000, max_width_to_hight_ratio=6): return is_large_enough(cont, min_area) and has_acceptable_format(cont, max_width_to_hight_ratio) def detect_figures(image: np.array): image = image.copy() image = remove_primary_text_regions(image) cnts = detect_large_coherent_structures(image) cnts = filter(is_likely_figure, cnts) rects = map(cv2.boundingRect, cnts) rects = remove_included(rects) return list(rects) def detect_figures_in_pdf(pdf_path, page_index=1, show=True): page = pdf2image.convert_from_path(pdf_path, first_page=page_index + 1, last_page=page_index + 1)[0] page = np.array(page) redaction_contours = detect_figures(page) page = draw_rectangles(page, redaction_contours) if show: show_mpl(page) else: return page