1

I have an image that contains repeated fainlty occuring patterns within it. I am interested in extracting only the patterns and turn other features in the image. Is there away to do this using python, especially, opencv or anyother python image library. I have seen the post here but it is not excatly what I am looking for. enter image description here

I have tried to convert the image from bgr to hsv, apply different filters like bilateral filter, histogram equalisation and combination of thresholding without any success.

_, thres_image = cv2.threshold(image, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)
image_ycrcb = cv2.CV.cvtColor(image, cv2.COLOR_RGB2YCrCb)
chal_y, _, _ = cv2.split(image_ycrcb)
equ_image = cv2.equalizeHist(chal_y)
BFAOZ
  • 29
  • 2
  • input data would help. -- none of those will do anything. you could try locally adaptive thresholding, or subtract a lowpass (gaussian, median, ...) https://i.stack.imgur.com/aGBST.png looks like some "imaginary property protection" watermark to me – Christoph Rackwitz May 18 '23 at 15:24

0 Answers0