Distributed source coding techniques for lossless compression of hyperspectral images

This paper deals with the application of distributed source coding (DSC) theoryto remote sensing image compression. Although DSC exhibits a significantpotential in many application fields, up till now the results obtained on realsignals fall short of the theoretical bounds, and often impose additionalsystem-level constraints. The objective of this paper is to assess the potentialof DSC for lossless image compression carried out onboard a remote platform. We first provide a brief overview of DSC of correlated information sources. Wethen focus on onboard lossless image compression, and apply DSC techniques inorder to reduce the complexity of the onboard encoder, at the expense of thedecoder's, by exploiting the correlation of different bands of a hyperspectraldataset. Specifically, we propose two different compression schemes, one based onpowerful binary error-correcting codes employed as source codes, and one based onsimpler multilevel coset codes. The performance of both schemes is evaluated on afew AVIRIS scenes, and is compared with other state-of-the-art 2D and 3D coders. Both schemes turn out to achieve competitive compression performance, and one of them also has reduced complexity. Based on these results, we highlight the mainissues that are still to be solved to further improve the performance ofDSC-based remote sensing systems.