每当我使用 pytesseract.image_to_string 它都会导致错误

Question

import pytesseract
import cv2
img = cv2.imread('ocrtest.jpg')   
img = cv2.resize(img, (600,400))
print(pytesseract.image_to_string(img))
cv2.imshow('Result', img)
cv2.waitKey(0)

Whenever I add line 5 - the print(pytesseract.image_to_string(img)) line - , it results in this giant error saying file not found error, and at the very bottom it says tesseract not found. I am using Anaconda Jupyter notebook, and pytesseract is downloaded in the Anaconda file, not the Jupyter notebook file on my desktop. Could that be the problem? If that line is not added, the code runs correctly and opens up the image file. The jpg image is downloaded within the desktop folder.

Answer 1

The ndarray from opencv is not support for pytesseract text extraction. Make use of pillow instead of opencv-python .

import pytesseract
from PIL import Image
img = Image.open("ocrtest.jpg")
resized_img = img.resize((600,400))
print(pytesseract.image_to_string(resized_img))
img.show()

Hope this will help you.