https://github.com/mingtaoguo/cnn-for-chinese-calligraphy-styles-classification

A convolution neural network with SE block and haar wavelet block for Chinese calligraphy styles classification by TensorFlow.(Paper: A novel CNN structure for fine-grained classification of Chinesecalligraphy styles)
https://github.com/mingtaoguo/cnn-for-chinese-calligraphy-styles-classification

cnn-classification haarwavelet squeeze-and-excitation tensorflow tensorflow-models

Last synced: 4 months ago
JSON representation

A convolution neural network with SE block and haar wavelet block for Chinese calligraphy styles classification by TensorFlow.(Paper: A novel CNN structure for fine-grained classification of Chinesecalligraphy styles)

• Host: GitHub
• URL: https://github.com/mingtaoguo/cnn-for-chinese-calligraphy-styles-classification
• Owner: MingtaoGuo
• Created: 2018-05-05T02:51:42.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2019-06-12T13:23:23.000Z (almost 6 years ago)
• Last Synced: 2025-01-31T03:01:52.334Z (4 months ago)
• Topics: cnn-classification, haarwavelet, squeeze-and-excitation, tensorflow, tensorflow-models
• Language: Python
• Homepage:
• Size: 9.67 MB
• Stars: 41
• Watchers: 2
• Forks: 4
• Open Issues: 2
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# CNN-for-Chinese-Calligraphy-Styles-classification

This code mainly solves the problem of style classification of Chinese calligraphy.

The paper has published: [A novel CNN structure for fine-grained classification of Chinesecalligraphy styles](https://rdcu.be/bxS19)

Introduce

----------

![fivefonts](https://github.com/MingtaoGuo/Simple-CNN-for-Chinese-character-classification/raw/master/IMGS/fivefonts.jpg)

There are four famous styles in standard font that is ou, yan, liu, zhao.

![fourstyles](https://github.com/MingtaoGuo/Simple-CNN-for-Chinese-character-classification/raw/master/IMGS/fourstyles.jpg)

Methods

---------

We use Squeeze-and-Excitation block and haar-wavelet block to improve the performance of our network.

### The method of Squeeze-and-Excitation block:

![SEBLOCK](https://github.com/MingtaoGuo/Simple-CNN-for-Chinese-character-classification/raw/master/IMGS/seblock.jpg)

### The method of haar-wavelet block:

![SEBLOCK](https://github.com/MingtaoGuo/Simple-CNN-for-Chinese-character-classification/raw/master/IMGS/haarwavelet.jpg)

### The architecture of our model:

![SEBLOCK](https://github.com/MingtaoGuo/Simple-CNN-for-Chinese-character-classification/raw/master/IMGS/network.jpg)