https://github.com/immortal13/EMS-GCN-hyperspectral-image-classification
Demo code of "EMS-GCN: An End-to-End Mixhop Superpixel-Based Graph Convolutional Network for Hyperspectral Image Classification"
https://github.com/immortal13/EMS-GCN-hyperspectral-image-classification
Last synced: 2 months ago
JSON representation
Demo code of "EMS-GCN: An End-to-End Mixhop Superpixel-Based Graph Convolutional Network for Hyperspectral Image Classification"
- Host: GitHub
- URL: https://github.com/immortal13/EMS-GCN-hyperspectral-image-classification
- Owner: immortal13
- Created: 2023-07-14T08:22:53.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2024-09-27T02:48:25.000Z (8 months ago)
- Last Synced: 2024-10-31T17:38:56.377Z (7 months ago)
- Language: C++
- Size: 27.3 KB
- Stars: 11
- Watchers: 2
- Forks: 1
- Open Issues: 3
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# EMS-GCN-hyperspectral-image-classification
Demo code of ["EMS-GCN: An End-to-End Mixhop Superpixel-Based Graph Convolutional Network for Hyperspectral Image Classification"](https://ieeexplore.ieee.org/document/9745164)
## Step 1: compiling cuda files
```
cd lib
. install.sh ## please wait for about 5 minutes
```
you can also refer to [ESCNet](https://github.com/Bobholamovic/ESCNet) for the compiling process.
## Step2: train and test
```
cd ..
CUDA_VISIBLE_DEVICES='7' python main.py
```
## Step3: record classification result
## Citation
If you find this work interesting in your research, please kindly cite:
```
@ARTICLE{9745164,
author={Zhang, Hongyan and Zou, Jiaqi and Zhang, Liangpei},
journal={IEEE Transactions on Geoscience and Remote Sensing},
title={EMS-GCN: An End-to-End Mixhop Superpixel-Based Graph Convolutional Network for Hyperspectral Image Classification},
year={2022},
volume={60},
number={},
pages={1-16},
doi={10.1109/TGRS.2022.3163326}}
```
Thank you very much! (*^▽^*)
This code is constructed based on [ESCNet](https://github.com/Bobholamovic/ESCNet) and [CEGCN](https://github.com/qichaoliu/CNN_Enhanced_GCN), thanks~💕.
If you have any questions, please feel free to contact me (Jiaqi Zou, [email protected]).