作者PHONm (USA~USA)
看板Python
标题[问题] 使用pytesseract 做ocr
时间Mon Aug 8 17:32:48 2016
我想要做字元辨识，但是字元图像有些破裂，有些字元会变成乱码，
所以就用OpenCV先进行一些前处理，然後存成新档後再进行一次OCR，
但是会有UnicodeDecodeError，可是程式码都没有用到中文啊@@!
不晓得是否是OpenCV转档那边出问题，
=====================Result=====================
<PIL.PngImagePlugin.PngImageFile image mode=RGBA size=397x112 at 0x3C0DF28>
24-D 1813f-ml 1-1
154?Dbb
<PIL.BmpImagePlugin.BmpImageFile image mode=RGB size=397x112 at 0x1131080>
Traceback (most recent call last):
File "C:/Users/cash.chien/PycharmProjects/OCR/OCRv1.1.py", line 19, in
<module>
str2 = image_to_string(img2)
File "C:\Anaconda3\lib\site-packages\pytesseract\pytesseract.py", line 167,
in image_to_string
return f.read().strip()
UnicodeDecodeError: 'cp950' codec can't decode byte 0xe2 in position 11:
illegal multibyte sequence
========================以下为原始码=======================
from pytesseract import image_to_string
from PIL import Image
import time
import cv2
import numpy as np
img = Image.open('12.bmp')
print(img)
str = image_to_string(img)
print(str)
img1 = cv2.imread('12.bmp',1)
kernel = np.ones((3,3))
opening = cv2.morphologyEx(img1, cv2.MORPH_OPEN, kernel)
cv2.imwrite('opening.bmp',opening)
img2 = Image.open('opening.bmp')
print(img2)
str2 = image_to_string(img2)
print(str2)
感谢!
--
※ 发信站: 批踢踢实业坊(ptt.cc), 来自: 59.124.131.189
※ 文章网址: https://webptt.com/cn.aspx?n=bbs/Python/M.1470648770.A.C41.html
1F：推 Sunal: 应该是cmd的输出问题 改成utf8试试 08/09 02:16
2F：推 goldflower: 先转str的编码 不过你直接把str命名掉不太好吧XD 08/11 00:22