Lossless Image Compression Based on a Generalized Recursive Interpolative DPCM

It is shown that the generalized recursive interpolation algorithm (GRINT) proposed is the most effective hierarchical technique for reversible compression of images that typically occur in remote sensing. The main advantage of the novel scheme with respect to other noncausal DPCM schemes is that interpolation is performed from all error-free values, thereby reducing the variance of residuals. Tests on LandSat, NOAA/AVHRR, MeteoSat, and SPOT images show that GRINT outperforms established hierarchical techniques, and also lossless JPEG and optimum DPCM when dealing with SPOT data, with the further advantage that GRINT makes error-free tokens available at any resolution, thereby expediting remote browsing on large image data-bases.