TU-Net: A Precise Network for Tongue Segmentation

by Yating Huang, Zhihui Lai, Wenjing Wang

Summary:

Intoduction:

This repository is for our ICCPR2020 paper "TU-Net: A Precise Network for Tongue Segmentation"

Framework:

Usage:

Requirement:

Ubuntu 16.04+pycharm+python3.6+pytorch1.7.1

Preprocessing:

Clone the repository:

git clone https://github.com/Yating-Huang/TU-Net.git
cd TU-Net

How to run:

The only thing you should do is enter the dataset.py and correct the path of the datasets. then run ~ example:

python main.py --action train&test --arch TUNet --epoch 2 --batch_size 2 

Results

after train and test,3 folders will be created,they are "result","saved_model","saved_predict".

saved_model folder:

After training,the saved model is in this folder.

result folder:

in result folder,there are the logs and the line chart of metrics.such as:

saved_predict folder:

in this folder,there are the ouput predict of the saved model,such as:

the datasets:

the Tongue dataset link：https://github.com/BioHit/TongeImageDataset

Citation:

If you found Triple ANet helpful for your research, please cite our paper:

@inproceedings{DBLP:conf/iccpr/HuangLW20,
  author    = {Yating Huang and
               Zhihui Lai and
               Wenjing Wang},
  title     = {TU-Net: {A} Precise Network for Tongue Segmentation},
  booktitle = {{ICCPR} 2020: 9th International Conference on Computing and Pattern
               Recognition, Xiamen, China, October 30 - Vovember 1, 2020},
  pages     = {244--249},
  publisher = {{ACM}},
  year      = {2020},
  url       = {https://doi.org/10.1145/3436369.3437428},
  doi       = {10.1145/3436369.3437428},
  timestamp = {Tue, 19 Jan 2021 15:37:23 +0100},
  biburl    = {https://dblp.org/rec/conf/iccpr/HuangLW20.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

Acknowledgement

Part of the code is revised from the UNET-ZOO.

Note

• The repository is being updated
• Contact: Yating Huang (huangyating2019@email.szu.edu.cn)