Tesseract OCR

本文详细介绍Tesseract OCR v5.0.0-alpha版本的基本使用方法,包括命令行参数配置、页面分割模式、语言包安装及自定义训练流程。同时,提供资源链接,帮助读者深入理解如何提升OCR识别质量,并介绍了EasyPR中文车牌识别系统。

摘要生成于 C知道 ,由 DeepSeek-R1 满血版支持, 前往体验 >

 

Tesseract基本使用

 

版本:

Tesseract OCR v5.0.0-alpha

Git Hub 地址:

https://github.com/tesseract-ocr/

语言包:

https://github.com/tesseract-ocr/tessdata

文档:

https://github.com/tesseract-ocr/docs

https://stackoverflow.com/questions/44619077/pytesseract-ocr-multiple-config-options

Page segmentation modes:

  0    Orientation and script detection (OSD) only.

  1    Automatic page segmentation with OSD.

  2    Automatic page segmentation, but no OSD, or OCR.

  3    Fully automatic page segmentation, but no OSD. (Default)

  4    Assume a single column of text of variable sizes.

  5    Assume a single uniform block of vertically aligned text.

  6    Assume a single uniform block of text.

  7    Treat the image as a single text line.

  8    Treat the image as a single word.

  9    Treat the image as a single word in a circle.

 10    Treat the image as a single character.

 11    Sparse text. Find as much text as possible in no particular order.

 12    Sparse text with OSD.

 13    Raw line. Treat the image as a single text line,

                        bypassing hacks that are Tesseract-specific.

基本命令演示:

tesseract filename output -l lang --psm 1
                       Tesseract OCR训练
Tesseract项目中的训练介绍

https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract-4.00

使用jTessBoxEditor来训练自己的数据 
https://www.jianshu.com/p/31afd7fc5813

https://www.cnblogs.com/cnlian/p/5765871.html

https://www.cnblogs.com/xpwi/p/9604567.html

采用这个工具遇到的问题(未解决)

 

案例相关

https://github.com/tesseract-ocr/tesseract/issues/1452

http://tpgit.github.io/UnOfficialLeptDocs/leptonica/index.html

改进质量

https://github.com/tesseract-ocr/tesseract/wiki/ImproveQuality

TensorFlow训练

https://github.com/EdjeElectronics/TensorFlow-Object-Detection-API-Tutorial-Train-Multiple-Objects-Windows-10#appendix-common-errors

 

 

EasyPR是一个开源的中文车牌识别系统

https://github.com/liuruoze/EasyPR

https://www.cnblogs.com/subconscious/p/3979988.html

 

其他公司的产品测试:

http://khmerocr.open.org.kh/api/ocr_image_to_text by providing input based-64 image.

http://khmerocr.open.org.kh/

Google

 

 

                                          编程语言

编程语: python
Tesseract: pytesseract

图像处理库:  OpenCV

 

评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值