python - Transform image to make characters readable

Question

I know the question have been asked many time but I'm at a lost with this image.

Does anyone have an idea how to transform this image so tesseract can correctly extract a string from the image.

As of now I tried some basic morpho transformations. I've also tried to remove the horizontal line.

This is my code:

import cv2
import pytesseract
import numpy as np

pytesseract.pytesseract.tesseract_cmd = r"C:Program FilesTesseract-OCResseract.exe"
net = cv2.dnn.readNet("frozen_east_text_detection.pb")
image = cv2.imread('captcha/captcha.png')

gray = cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)
thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]

# Remove horizontal
horizontal_kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (25,1))
detected_lines = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, horizontal_kernel, iterations=2)
cnts = cv2.findContours(detected_lines, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if len(cnts) == 2 else cnts[1]
for c in cnts:
    cv2.drawContours(image, [c], -1, (255,255,255), 2)

# Repair image
repair_kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (1,6))
result = 255 - cv2.morphologyEx(255 - image, cv2.MORPH_CLOSE, repair_kernel, iterations=1)
kernel = np.ones((2,2),np.uint8)
result = cv2.morphologyEx(result, cv2.MORPH_CLOSE, kernel)


cv2.imshow('thresh', thresh)
cv2.imshow('detected_lines', detected_lines)
cv2.imshow('image', image)
cv2.imshow('result', result)
cv2.waitKey()

txt = pytesseract.image_to_string(result, lang='eng', config="--psm 7 --oem 3 -l eng")

print(txt)

Answer 1

等待大神答复