Awesome

Landslide mapping on SAR data by Attention-U-Net

This repository contains codes and sample data from the journal paper "Rapid Mapping of landslide on SAR data by Attention U-net" by Nava et al (2022). With this repository files you can train the Attention U-Net segmentation model on Sentinel-1 SAR amplitude landslide data sampled for the Hokkaido multiple landslide event of September 2018.

Contact emails: lorenava996@gmail.com & kushanavb@gmail.com

Attention U-Net architecture