M-UNet

This is a official implementation for our paper: "Simple Multiscale UNet for Change Detection With Heterogeneous Remote Sensing Images" has been published on IEEE GEOSCIENCE AND REMOTE SENSING LETTERS by Zhiyong Lv, Haitao Huang, Lipeng Gao, Jón Atli Benediktsson, Minghua Zhao and Cheng Shi.

Requirements

python=3.8 pytorch=1.9
opencv-python=4.6.0.66
scikit-image=0.18.1
scikit-learn=0.24.1

Usage

Train

Load the train and test(val) data path
python train.py

Test

Load the model path
Load the test data path
python test.py

Citation

If you find our work useful for your research, please consider citing our paper:

@ARTICLE{9770788,
  author={Lv, Zhiyong and Huang, Haitao and Gao, Lipeng and Benediktsson, Jón Atli and Zhao, Minghua and Shi, Cheng},
  journal={IEEE Geoscience and Remote Sensing Letters}, 
  title={Simple Multiscale UNet for Change Detection With Heterogeneous Remote Sensing Images}, 
  year={2022},
  volume={19},
  number={},
  pages={1-5},
  doi={10.1109/LGRS.2022.3173300}}

Contact us

If you have any problme when running the code, please do not hesitate to contact us. Thanks.
E-mail: Lvzhiyong_fly@hotmail.com Date: Mar 17, 2023