0

I'm trying to OCR image with pytesseract. Once I do the OCR for below image the result shows as "WV over"

What are the image pre-processing techniques that can be use to enhance this image by filling missing parts of text.

text_image

Enhance image OCR ability by,

  • fill missing text parts in image

  • Improve quality of image

Please let me know sample code that can be used to fulfil above requirements.

STF
  • 1,485
  • 3
  • 19
  • 36
Chinthaka
  • 9
  • 3

1 Answers1

0

can you try morphological operations like dilation and see if it improves ?

balu
  • 1,023
  • 12
  • 18
  • Thank you for your respond. I have try dilation but it is not working as expected. kernel = np.ones((5,5), np.uint8) image = cv2.dilate(image, kernel, iterations = 1) – Chinthaka Feb 09 '23 at 10:24