Satellite on-board image compression adviser

The next generation of earth observation satellites will exceed their downlink capability making essential to compress images on-board the satellites. Due to the limited download bandwidth, the limited contact time between the satellite and the ground systems and the huge data volume captured by the new high resolution sensors, efficient image compression systems must be included on observation satellites. These systems will be very often based on the JPEG2000 standard, which has proved to be one of the most powerful current image compression algorithms. It is based on the discrete wavelet transform and allows lossless and lossy compression. We present a compression adviser, a system capable of analyze images statistical features to decide what kind of compression, lossless or lossy, is more profitable to use in terms of storage and bandwidth resources utilization. A first FPGA prototype of the adviser coprocessor is presented to evaluate its design and performance.