ERROR-RESILIENT PYRAMID VECTOR QUANTIZATION FOR IMAGE COMPRESSION

Pyramid vector quantization (PVQ) uses the lattice points of a pyramid al shape in multidimensional space as the quantizer codebook, It is a fixed-rate quantization technique that can be used for the compression of Laplacian-like sources arising from transform and subband image co ding, where its performance approaches the optimal entropy-coded scala r quantizer without the necessity of variable length codes. In this pa per, we investigate the use of PVQ for compressed image transmission o ver noisy channels, where the fixed-rate quantization seduces the susc eptibility to bit-error corruption. We propose a new method of derivin g the indices of the lattice points of the multidimensional pyramid an d describe how these techniques can also improve the channel noise imm unity of general symmetric lattice quantizers. Our new indexing scheme improves channel robustness by up to 3 dB over previous indexing meth ods, and can be performed with similar computational rest. The final f ixed-rate coding algorithm surpasses the performance of typical Joint Photographic Experts Group (JPEG) implementations and exhibits much gr eater error resilience.