Created
March 21, 2023 08:25
-
-
Save nopium/801405674855926555c8a3efa288e7af to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import cv2 | |
| import numpy as np | |
| from pdf2image import convert_from_path | |
| import argparse | |
| parser = argparse.ArgumentParser() | |
| parser.add_argument("filename", help="Enter a pdf filename to process") | |
| args = parser.parse_args() | |
| def dilate_image(image): | |
| full_width = 330 | |
| check_width = 110 | |
| crop = image[610:610+1630, 440:440+check_width] | |
| #cv2.imwrite("crop.png", crop) | |
| gray = cv2.cvtColor(crop, cv2.COLOR_BGR2GRAY) | |
| #cv2.imwrite("gray.png", gray) | |
| blur = cv2.GaussianBlur(gray, (7,7), 0) | |
| #cv2.imwrite("blur.png", blur) | |
| thresh = cv2.threshold(blur, 0,255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1] | |
| #cv2.imwrite("thresh.png", thresh) | |
| kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (3,13)) | |
| #cv2.imwrite("kernel.png", kernel) | |
| dilate = cv2.dilate(thresh, kernel, iterations=1) | |
| # cv2.imwrite("dilate.png", dilate) | |
| return dilate | |
| def detect_values(dilate, deviance = 1.1): | |
| # loop rows | |
| start_y = 0 | |
| row_height = 160 | |
| hist = [] | |
| for i in range(1, 11): | |
| row_img = dilate[ start_y:start_y + row_height, 0:dilate.shape[1] ] | |
| # cv2.imwrite('row-%d.png' % i, row_img) | |
| start_y = start_y + row_height | |
| white_pixels = cv2.countNonZero(row_img) | |
| total_pixels = row_img.size | |
| avg_pixels = white_pixels / total_pixels | |
| hist.append(avg_pixels) | |
| mean = np.min(hist) | |
| diff_from_mean = abs(hist - mean) | |
| threshold = deviance * np.std(hist) | |
| deviant_indices = np.where(diff_from_mean > threshold)[0] | |
| ret = [] | |
| for index in deviant_indices: | |
| # print(index+1, hist[index]) | |
| ret.append(index+1) | |
| return ret | |
| pages = convert_from_path(args.filename, 500) | |
| for num,page in enumerate(pages): | |
| # page.save('out-%d.png' % num, 'PNG') | |
| image = np.array(page) | |
| dilate = dilate_image(image) | |
| values = detect_values(dilate) | |
| print(f'Page: {num+1} checked: {values}') | |
| # Convert RGB to BGR | |
| #open_cv_image = open_cv_image[:, :, ::-1].copy() | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment