On-board compression algorithm for satellite multispectral images

Summary form only given. This paper proposes a low complexity and flexible fixed data rate compression algorithm for multispectral imagery specifically targeted for high-rate instruments used onboard of spacecraft. To reduce the development costs, this algorithm re-use, as far as possible, already existing compressors, such as CNES algorithms already used in previous missions or standard algorithms suited for space missions. The proposed algorithm is then composed of a decorrelation stage performed on spectral channels and can be used in front of existing monospectral compressors. The decorrelation stage is based either on a fixed transform evaluated previously on a set of images or on the Karhunen-Loeve transform locally adapted to the studied block of image, implemented in a scan-based mode