> ## Content Index
> Fetch the complete content index at: https://cms.yufeiran.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# CNN汉字识别
- URL: https://cms.yufeiran.com/yi-zi-shi-bie-ren-gong-zhi-neng-da-zuo-ye/
- Published: 2020-05-25T04:33:38.000Z
- Updated: 2026-09-28T10:56:11.000Z
- Author: Feiran Yu
- Tags: 作品, #legacy-import

## 目标

训练CNN网络，识别常用汉字

## 依赖环境

python 2.X(推荐2.7）  
tensorflow 1.1.15

## 用法

1. 配置运行环境：安装python2.x ，tensorflow 1.1.15
2. 克隆[https://github.com/yufeiran/ChineseRecognizer](https://github.com/yufeiran/ChineseRecognizer?ref=cms.yufeiran.com)  
到本地目录
3. \[src\] 为图片路径  
`python main.py [src]`

## 参考链接

[【OCR技术系列之四】基于深度学习的文字识别（3755个汉字）](https://www.cnblogs.com/skyfsm/p/8443107.html?ref=cms.yufeiran.com)  
[用TENSORFLOW实现一个手写汉字识别之一：数据的获取和处理](http://www.rethink.fun/index.php/2018/05/03/hand%5Fwriting%5Frecgnition/?ref=cms.yufeiran.com)

## GitHub仓库

完整汉字识别程序  
[ChineseRecognizer](https://github.com/yufeiran/ChineseRecognizer?ref=cms.yufeiran.com)  
深度学习入门的书的学习笔记  
[DeepLearn](https://github.com/yufeiran/DeepLearn?ref=cms.yufeiran.com)

## 参考书籍

[《深度学习入门》](https://book.douban.com/subject/30270959//?ref=cms.yufeiran.com)