https://github.com/ihsandevs/captcha-recognizer

https://github.com/ihsandevs/captcha-recognizer

Last synced: 6 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/ihsandevs/captcha-recognizer
• Owner: IhsanDevs
• License: apache-2.0
• Created: 2023-11-26T20:15:13.000Z (12 months ago)
• Default Branch: main
• Last Pushed: 2023-11-26T20:15:16.000Z (12 months ago)
• Last Synced: 2023-11-26T22:23:45.085Z (12 months ago)
• Language: Python
• Size: 22.5 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
Introduce

=========

Image recognition captchas using TensorFlow, no need image segmentation, run on Ubuntu 18.04, Python 3.10, Tensorflow 2.10.0, Numpy 1.23.4

![captcha](https://raw.githubusercontent.com/PatrickLib/captcha_recognition/master/data/test_data/1ab2s_num286.jpg)![captcha](https://raw.githubusercontent.com/PatrickLib/captcha_recognition/master/data/test_data/1ezx8_num398.jpg)![captcha](https://raw.githubusercontent.com/PatrickLib/captcha_recognition/master/data/test_data/1iv22_num346.jpg)![captcha](https://raw.githubusercontent.com/PatrickLib/captcha_recognition/master/data/test_data/1kxw2_num940.jpg)![captcha](https://raw.githubusercontent.com/PatrickLib/captcha_recognition/master/data/test_data/3mtj9_num765.jpg)![captcha](https://raw.githubusercontent.com/PatrickLib/captcha_recognition/master/data/test_data/1vuy5_num17.jpg)

Accuracy is directly related to generated Training data size and the Learning steps number.

For example, 

captcha generator: https://github.com/Gregwar/CaptchaBundle

 

Dependence

==========

### python 2.7

### Anaconda2 4.3.1

https://www.continuum.io/downloads#linux

### TensorFlow 1.1

https://github.com/tensorflow/tensorflow

### captcha

https://pypi.python.org/pypi/captcha/0.1.1

Usage

=====

## 1.prepare captchas

put your own captchas in **/data/train_data/** for training, **/data/valid_data/** for evaluating and **/data/test_data/** for recognize testing, images file name must be **label_\*.jpg** or **label_\*.png** and recommend size **128x48**. you can also use default generation:

```

python captcha_gen_default.py

```

## 2.convert dataset to tfrecords

the result file will be **/data/train.tfrecord** and **/data/valid.tfrecord**

```

python captcha_records.py

```

## 3.training

train and evaluate neural network on CPU or one single GPU

```

python captcha_train.py

```

you can also train over multiple GPUs

```

python captcha_multi_gpu_train.py

```

## 4.evaluate

```

python captcha_eval.py

```

## 5.recognize

read captchas from **/data/test_data/** for recogition

```

python captcha_recognize.py

```

result like this

```

...

image WFPMX_num552.png recognize ----> 'WFPMX'

image QUDKM_num468.png recognize ----> 'QUDKM'

```