AWD-Net: Hyperspectral and Multispectral Image Fusion Based on Adaptive Wavelet Transform and Dual Spectral-Spatial Branch

This is the official code of AWD-Net: Hyperspectral and Multispectral Image Fusion Based on Adaptive Wavelet Transform and Dual Spectral-Spatial Branch.

Network Architecture

1. Environment Setup:

1.1 Requirements

• Python >=3.8 (Recommended:Anaconda)
• NVIDIA GPU + CUDA

1.2 Create COnda Environment

conda create -n AWD python=3.8 -y
conda activate AWD
pip install -r requirements.txt

2. Data Preparation:

• Download Datasets(PaviaU.mat\Pavia.mat\Urban.mat\Washington_DC.mat)

• Place the Datasets to /AWD-Net/data/.

• Then this repo is collected as the following form:

  |--AWD-Net
      |--data  
              |--Pavia.mat
              |--PaviaU.mat
              |--Urban.mat
              |--Washington_DC.mat	

3.Training

python main.py

4.Testing

(1) Download the pretrained model zoo from and place them to /AWD-Net-main/checkpoints/.

(2) python test.py

Acknowledgment

This repo is based on SSRNET which is an exellent work.