简体   繁体   English

Pytesseract 输出与图片中的文字完全不同

[英]Pytesseract output is completely different from the text in the image

The following code runs on my friend's windows laptop perfectly but on mine, it spits out complete garbage.以下代码在我朋友的 Windows 笔记本电脑上完美运行,但在我的笔记本电脑上,它吐出完全垃圾。 I think it might have something to do with the libraries but I can't say for sure?我认为这可能与图书馆有关,但我不能肯定? Can you guys let me know if it runs for you or if you have ever had this happen before?你们能告诉我它是否适合你,或者你以前是否遇到过这种情况?

import pytesseract
from PIL import Image, ImageEnhance, ImageFilter
from textblob import TextBlob

originalImage = Image.open("pictures\hp.png")
workingImage = originalImage.copy()
workingImage = originalImage.filter(ImageFilter.MedianFilter())

enhanceSharpness = ImageEnhance.Sharpness(workingImage)
enhanceColor = ImageEnhance.Color(workingImage)

workingImage = enhanceSharpness.enhance(2)
workingImage = enhanceColor.enhance(0)
workingImage = workingImage.convert('1')

text = pytesseract.image_to_string(workingImage)
correctedText = TextBlob(text).correct()

print(correctedText)

This is the picture we are feeding it:这是我们喂它的图片: 在此处输入图片说明

Rather than printing the text in the photo as it does on my friend's computer, mine prints the following:我没有像在我朋友的电脑上那样打印照片中的文字,而是打印以下内容:

of was an am: came: mm: mmm mm bu noma of arm 523w at"
mmmhmg peaches? w 3 cm {mamma a map m: K wm} Or Mummy
didn‘c realm when Am had cm then he mmm Am: Ma mad m

Am!» adam Them was 3 W317); Ll them on arm mmm a‘met
Turn mm "mm wasn't a mm: m 313% Ham sum Am have. term mmm
M‘) It must had: {mm 3 mm fiftht‘ High: Or Him‘cy [asked and
hated :31 me cm of Taxed back Of I Mummy time mound! m:

come and up in: may M maxim} tin: ca: m Am mm‘m h Am raw
made m: mm mm mid and Run“ w Of iwkmg at me 5 Is
cmricbfl mad may or $$ka Or [)unity gave hrmsuifa hut: wake and
put 11% cm my m“ Am Am} Is he rim mere own M {high W
msihmg except & Large may of'dry‘b M. M h my m E“ that thy

I downloaded tesseract-ocr version 4.0 and it works perfectly now.我下载了 tesseract-ocr 4.0 版,现在运行良好。 Thanks for all your help guys!感谢您的帮助!

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM