Adjustable Spatio-Spectral Hyperspectral Image Compression Network
Martin Hermann Paul Fuchs, Behnood Rasti, and Begüm Demir · 2025
Read Adjustable Spatio-Spectral Hyperspectral Image Compression Network by Martin Hermann Paul Fuchs, Behnood Rasti, and Begüm Demir (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-23447.txt.
Description
Martin Fuchs and co-authors present a flexible network for squeezing hyperspectral images, in a 2025 paper. Hyperspectral images record many light bands per pixel, so archives grow fast. Learned compression helps, but how squeezing the color bands and the spatial size interact was unclear. Their network, HyCASS, lets users adjust each dimension on its own.
Six modules handle spectral coding, spatial coding, and rate control on both ends. Layers mix local filters with transformer blocks to catch both near and far repeats in the data. Tests on three benchmark data sets beat prior learned methods. A single adjustable model replaces several fixed ones. Remote sensing archives can store more scenes at the same quality.
Catalog record
| CREATOR | Martin Hermann Paul Fuchs, Behnood Rasti, and Begüm Demir |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-23447