Adaptive Classified Pattern Matching Vector Quantization Approach For Compressing Images

In this paper, we propose a new, two-stage, adaptive classified pattern-matching algorithm based on residual vector quantization and variable bit rate encoding techniques as they have proved to be very efficient in progressive image transmission and compression rate. Both the intra-block and inter-block correlation properties have been exploited to enhance the compression rate. The innovative aspects of this method are: i) the use of high eigenvalued blocks as initial seeds yielding fast convergence; ii) the use of number patterns as the secondary codebook; and, iii) the generation of number patterns using basic combinatorics without requiring complex calculations; iv) use of variable number of indices.