Adjustable Spatio-Spectral Hyperspectral Image Compression Network

TL;DR

This paper introduces HyCASS, a novel learning-based hyperspectral image compression network that allows adjustable compression in spectral and spatial domains, improving compression performance and providing guidelines for balancing spectral and spatial redundancies.

Contribution

We propose HyCASS, a flexible, learning-based HSI compression model with adjustable spectral and spatial compression, employing transformer-based modules for redundancy exploitation.

Findings

HyCASS outperforms existing models by up to 2.36 dB PSNR.

The model effectively balances spectral and spatial compression across different ratios.

Experimental results validate the model's adaptability and superior compression quality.

Abstract

With the rapid growth of hyperspectral data archives in remote sensing (RS), the need for efficient storage has become essential, driving significant attention toward learning-based hyperspectral image (HSI) compression. However, a comprehensive investigation of the individual and joint effects of spectral and spatial compression on learning-based HSI compression has not been thoroughly examined yet. Conducting such an analysis is crucial for understanding how the exploitation of spectral, spatial, and joint spatio-spectral redundancies affects HSI compression. To address this issue, we propose Adjustable Spatio-Spectral Hyperspectral Image Compression Network (HyCASS), a learning-based model designed for adjustable HSI compression in both spectral and spatial dimensions. HyCASS consists of six main modules: 1) spectral encoder module; 2) spatial encoder module; 3) compression ratio (CR) adapter encoder module; 4) CR adapter decoder module; 5) spatial decoder module; and 6) spectral decoder module. The modules employ convolutional layers and transformer blocks to capture both short-range and long-range redundancies. Experimental results on three HSI benchmark datasets demonstrate the effectiveness of our proposed adjustable model compared to existing learning-based compression models, surpassing the state of the art by up to 2.36 dB in terms of PSNR. Based on our results, we establish a guideline for effectively balancing spectral and spatial compression across different CRs, taking into account the spatial resolution of the HSIs. Our code and pre-trained model weights are publicly available at https://git.tu-berlin.de/rsim/hycass .