Stirling-PDF/scripts/detect-blank-pages.sh
Anthony Stirling 0b1cdf6a68 remove blanks
2023-05-08 12:18:48 +01:00

39 lines
No EOL
1.2 KiB
Bash

import cv2
import numpy as np
import sys
def is_blank_image(image_path, threshold=10, white_value=255, blur_size=5):
image = cv2.imread(image_path, cv2.IMREAD_GRAYSCALE)
if image is None:
print(f"Error: Unable to read the image file: {image_path}")
return False
# Apply Gaussian blur to reduce noise
blurred_image = cv2.GaussianBlur(image, (blur_size, blur_size), 0)
_, thresholded_image = cv2.threshold(blurred_image, white_value - threshold, white_value, cv2.THRESH_BINARY)
# Calculate the percentage of white pixels in the thresholded image
white_pixels = np.sum(thresholded_image == white_value)
total_pixels = thresholded_image.size
white_pixel_percentage = (white_pixels / total_pixels) * 100
return white_pixel_percentage > 99
if __name__ == "__main__":
if len(sys.argv) != 2:
print("Usage: python detect_blank_page.py <image_path>")
sys.exit(1)
image_path = sys.argv[1]
blank = is_blank_image(image_path)
if blank:
# Return code 1: The image is considered blank.
sys.exit(1)
else:
# Return code 0: The image is not considered blank.
sys.exit(0)