2

How do I isolate or crop only the handwritten text using OpenCV and Phyton for the image:

Handwritten image

I have tried to use:

cv2.findContours

but because of the noise (background and dirty in paper) I can't get only the paper.

How do I do this?

gudok
  • 4,029
  • 2
  • 20
  • 30
Sergio Pantano
  • 111
  • 1
  • 10

2 Answers2

2

To smooth noisy images, typical methods are to apply some type of blurring filter. For instance cv2.GaussianBlur(), cv2.medianBlur(), or cv2.bilaterialFilter() can be used to remove salt/pepper noise. After blurring, we can threshold to obtain a binary image then perform morphological operations. From here, we can find contours and filter using aspect ratio or contour area. To crop the ROI, we can use Numpy slicing

Detected text

enter image description here

Extracted ROI

enter image description here

Code

import cv2

image = cv2.imread('1.jpg')
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
blur = cv2.medianBlur(gray, 5)
thresh = cv2.adaptiveThreshold(blur,255,cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY_INV,11,8)

kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (5,5))
dilate = cv2.dilate(thresh, kernel, iterations=6)
cnts = cv2.findContours(dilate, 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:
    x,y,w,h = cv2.boundingRect(c)
    ROI = image[y:y+h, x:x+w]
    cv2.imwrite('ROI.png', ROI)
    break

cv2.imshow('thresh', thresh)
cv2.imshow('dilate', dilate)
cv2.imshow('ROI', ROI)
cv2.waitKey()
nathancy
  • 42,661
  • 14
  • 115
  • 137
  • I did not work for me. I copied your code but the ROI prints the top of the image, the part above of paper. – Sergio Pantano Sep 27 '19 at 23:36
  • Using the input image you provided, I'm getting the correct result. Is your input image a .png or .jpg? Try with the updated code. The result should be the same – nathancy Sep 28 '19 at 00:32
  • I image is jpeg. Here is the ROI: [ROI](https://imgur.com/O4nsxAl) and the three together: [three](https://imgur.com/flWXsmq). Thanks for help. – Sergio Pantano Sep 28 '19 at 02:12
  • Your dilate image is different from mine. Very weird. Are you sure you have `cv2.MORPH_RECT, (5,5)` and `iterations=6`? Try change `iterations=6` to `iterations=15` – nathancy Sep 28 '19 at 02:14
1
  1. Convert image to single channel gray.
  2. Apply adaptiveThreshold to your image. Handwriting will become of black color, rest will be white.
  3. If you want to get segmentation for this word as a solid thing, then also apply morphologyEx, with MORPH_CLOSE. Here you should play with kernel, most likely it will be ellipse 3x3, and number of iterations, usually 5-10 iterations is ok.
   kernel = cv2.getStructuringElement(shape=cv2.MORPH_ELLIPSE, ksize=(3, 3))
   image = cv2.morphologyEx(image, cv2.MORPH_CLOSE, kernel, iterations=7)
  1. Use connectedComponentsWithStats. It will put every char into separate component. stats will hold bounding boxes either for whole word, or (if you omit step #2) it will hold info for each connected characters group.

P.S.: Let me know if you need full code example.

Stepan Dyatkovskiy
  • 813
  • 9
  • 15