0

I want to remove background and sharpen the images of following type: image 1

image 2

Both of these are signatures. I want to be able to remove everything except the signature itself and sharpen the lines of signature. I am able to get a mask using Canny edge detection using following code

import cv2
im_path  = r'test2.png'
image = cv2.imread(im_path) #args["image"]
image = cv2.resize(image, (680, 460))
#rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]

cv2.imshow('thresh', thresh) ###for showing 

cv2.imwrite('thresh.jpg', thresh) ###for saving
cv2.waitKey()

And these are masks I get; mask

But Im clueless about what Image processing operations to perform next.

PS: These signatures are same (not forged) and next step would be to find similarity between them.

Fatima Arshad
  • 119
  • 1
  • 9
  • You don’t have a Canny edge detection in that code. What is your goal?Do you want to set the background to be all white? What must happen to the blurriness at the edges of the signature in the input image? What is the intended use of the output? – Cris Luengo Jan 14 '21 at 15:15
  • @CrisLuengo I want to be able to match two images. Nothing fancy. Just need to be able to tell if two signatures are similar or not. – Fatima Arshad Jan 15 '21 at 12:06

1 Answers1

1

Try this

import cv2
import numpy as np

image = cv2.imread("r'test2.png'")
original = image.copy()
mask = np.zeros(image.shape, dtype=np.uint8)
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (5,5))
opening = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel, iterations=3)

cnts = cv2.findContours(opening, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if len(cnts) == 2 else cnts[1]
cnts = sorted(cnts, key=cv2.contourArea, reverse=True)
for c in cnts:
    cv2.drawContours(mask, [c], -1, (255,255,255), -1)
    break

close = cv2.morphologyEx(mask, cv2.MORPH_CLOSE, kernel, iterations=4)
close = cv2.cvtColor(close, cv2.COLOR_BGR2GRAY)
result = cv2.bitwise_and(original, original, mask=close)
result[close==0] = (255,255,255)

cv2.imshow('result', result)
cv2.waitKey()
Rahul Ramesh
  • 90
  • 7
  • https://stackoverflow.com/questions/58613825/get-boundary-from-canny-edges-and-remove-the-background-of-an-image – Rahul Ramesh Jan 14 '21 at 11:17
  • 1
    What does this do? Why is it supposed to help? Answers that teach something are the more valuable ones. An answer with just code helps no-one. – Cris Luengo Jan 14 '21 at 15:13
  • The answer he linked had more context. I think to say it helps no one is an overstatement, but @RahulRamesh it would be good to either just make this into a comment with the link *or* add some description to the answer and perhaps customize the code a bit (looks identical to the linked answer except for the filename). – Hack-R Feb 18 '22 at 17:00