[Python] How to perform OCR for several contours

I have a code that identifies contours in a licence plates, however I don't know how to extract the letters using pytesseract for each individual contour. This is the original image:



This is the code:

import cv2
import numpy as np
import pytesseract


image = cv2.imread('c1.png')
cv2.waitKey(0)


gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)


edged = cv2.Canny(gray, 30, 200)
cv2.waitKey(0)
blured = cv2.blur(gray, (5,5), 0)
img_thresh = cv2.adaptiveThreshold(blured, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY_INV, 11, 2)
rect_kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (30, 10))
threshed = cv2.morphologyEx(img_thresh, cv2.MORPH_CLOSE, rect_kernel)

contours, hierarchy = cv2.findContours(edged,
cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)
for contour in contours:
if cv2.contourArea(contour) > 170:
[X, Y, W, H] = cv2.boundingRect(contour)
cv2.rectangle(image, (X, Y), (X + W, Y + H), (0,0,255), 2)

mask = np.zeros(gray.shape,np.uint8)
new_image = cv2.drawContours(mask,[contour],0,255,-1,)
new_image = cv2.bitwise_and(image,image,mask=mask)
(x, y) = np.where(mask == 255)
(topx, topy) = (np.min(x), np.min(y))
(bottomx, bottomy) = (np.max(x), np.max(y))

Cropped0 = gray[topx:bottomx+2, topy:bottomy+2]


cv2.imshow('Canny Edges After Contouring', edged)
cv2.waitKey(0)


print("Number of Contours found = " + str(len(contours)))
result_number = pytesseract.image_to_string(new_image, lang='eng')
print("Detected Number is:",result_number)

cv2.imshow('Contours', image)
cv2.waitKey(0)
cv2.imshow('new_image', Cropped0)
cv2.waitKey(0)
cv2.destroyAllWindows()


The output of the code:



I tried cropping the image after identifying the contours for the OCR but it only crops number 5 for some reason. Also I would like to ignore the contours containing Arabic letters would it be possible to exclude these two from the recognition process?. Can someone please help me on this.

Continue reading...