I would like to compare two web pages images using computer vision techniques. Show what are non-unique portions comparing both images. Which part image1 not exist in image2 vice versa.
$\begingroup$
$\endgroup$
Add a comment
|
$\begingroup$
$\endgroup$
4
A simple and time efficent solution is the computation of the absoulte values of the difference image. This can be done as follows:
import cv2
img_a = cv2.imload([first_image_filename])
img_b = cv2.imload([second_image_filename])
img_diff = cv2.absdiff(img_a, img_b)
cv2.imshow('abs difference', img_diff)
cv2.waitKey(-1)
-
$\begingroup$ Thank you I can see the difference. Is it possible to ignore the same alphabets with different font sizes(Can we show as no difference)? $\endgroup$ Apr 29 '20 at 19:19
-
$\begingroup$ With such a low level analysis not. To ignore alphabets you would have to apply an optical character recognition (OCR) and than set regions with detected characters to 0. $\endgroup$ May 1 '20 at 8:48
-
$\begingroup$ Thank you, I will try to apply OCR and is there any similar kind of work or any blog. $\endgroup$ May 1 '20 at 14:19
-
$\begingroup$ I have got no specific blog in mind. Maybe learnopencv.com/tag/ocr $\endgroup$ May 1 '20 at 14:41