# BoundaryNets

**Repository Path**: zhanglibingo/boundary-nets

## Basic Information

- **Project Name**: BoundaryNets
- **Description**: This implementation is for the article titled "Cloud detection with BoundaryNets".
- **Primary Language**: Python
- **License**: GPL-3.0
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 3
- **Created**: 2022-06-10
- **Last Updated**: 2022-06-10

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Cloud detection with BoundaryNets
This implementation is for the article titled "Cloud detection with BoundaryNets".

-------------

## Environemnt
### Our environment
- CentOS 7.9
- Intel Xeon Gold 5218R CPU
- 2 NVIDIA Geforce 2080Ti GPUs
### Requirements
- Python 3.6
- Pytorch 0.4.1
- Gdal 
------

## Train
### Trained model

The trained model can be downloaded from: [Google Driver](https://drive.google.com/file/d/1t_YE6AR_2sbdjkccMu3zO7IsLwnOGWEd/view?usp=sharing).

### Training details
1. config the training details in the file: `boundarynets_train_sparcs.py`
2. train the nets by:
    ```shell
    python boundarynets_train_sparcs.py
    ```
-----
## Test
1. config the model_dir and saved_dir on the file: `boundarynets_test_sparcs.py`
2. inference the nets by:
    ```shell
    python boundarynets_test_sparcs.py
    ```

-------
## Todo
- [x] Test Code release
- [x] Trained model release
- [x] Train Code release
- [ ] Evalution Code release

-----
## BibTex
```
@article{WU2022218,
title = {Cloud detection with boundary nets},
journal = {ISPRS Journal of Photogrammetry and Remote Sensing},
volume = {186},
pages = {218-231},
year = {2022},
issn = {0924-2716},
doi = {https://doi.org/10.1016/j.isprsjprs.2022.02.010},
url = {https://www.sciencedirect.com/science/article/pii/S0924271622000521},
author = {Kang Wu and Zunxiao Xu and Xinrong Lyu and Peng Ren}
}
```