10-19-2018 01:32 PM
My problem is to detect whether the page is blank or not. The approach I have followed is to remove Gaussian noise and apply Canny edge detection. The last step I would like to do is remove the border . A hacky solution i thought of is to just crop the image but it there are some cases which will border in between. Could you share any ideas you have to remove the borders/lines?
im_gray = cv2.imread(curFile, cv2.IMREAD_GRAYSCALE)
blur = cv2.GaussianBlur(im_gray,(5,5),0)
edges = cv2.Canny(blur,100,200)
example image: https://www.dropbox.com/s/7zjpvlg42tsunxi/nlin.jpg?dl=0
10-22-2018 04:34 PM
Hey,
What software are you using to do this? I think there are a few options but curious to see what software and hardware you are using. Depending on the software, you could try to implement a Golden Template type approach. But again, I would like to know your setup.
Thanks,
-BrianJ.
04-01-2019 12:57 AM
for detection of a blank page, you have to follow a grayscale methodology, check your code again and change the input values.