Research on cultural-based multi-objective particle swarm optimization in image compression quality assessment

摘要

Quantization is a main factor which affects the performance of the JPEG compression. Compression ratio and the quality of the decoded images are both determined by quantization tables. Appropriate choice of the quantization tables is the key to the Compression performance. To select different optimal quantization tables for different classes of images is multi-objective optimal problem. A cultural-based multi-objective particle swarm optimization model is proposed in this paper. And, different trade-offs between image compression and quality is presented. The simulation result shows that the proposed model is effective to the choice of image compression quality.