Band ordering in compression of ultraspectral images - Springer

摘要

Atmospheric InfraRed Sounder images are a type of ultraspectral images that are good candidates to compression as they include several thousand bands that account for well over 40 MB of data per single cube. As part of most prediction based compression schemes, there is a preprocessing stage in which a reversible band ordering process is included to maximize spectral band correlation. In this paper, we analyze, compare, and propose novel improvements to different band ordering and correlation estimation techniques in the context of ultraspectral image compression. In order to measure the performance of these methods, we also introduce a new metric based on comparing a given sequence of spectral bands to the corresponding optimal one. Finally, we apply the techniques presented in this paper to a real compression algorithm in order to obtain compression rates as a function of band ordering and correlation factors while determining the best scenario.