Band Reordering Heuristic for Lossless Satellite Image Compression with CCSDS

Remote sensing satellite images are used widely in space imaging applications by acquiring significant information of earth surface. The size of these images is typically huge in amount and they need to transmit to the ground for a specific application. Thus, the efficient compression techniques are required to provide better bandwidth utilization for reducing transmission time. Since the images are actually strongly correlated spatially, spectrally and temporally, these correlations give ample opportunities to compress the data in various domains. In addition, the data features have a strong similarity in the disjoint spectral regions. Consequently, the similarity measurement based band reordering strategy is used for increasing the compression performance. However, the optimal band reordering is still a computationally challenging problem. In this paper, a correlation based heuristic has been proposed for the band reordering along with the recommended standard by Consultative Committee for Space Data Systems (CCSDS) 123 lossless encoder for better compression. The experimental result with real satellite images gives around 4.5-6.5% higher compression ratio than that of without reordering.