The computer desktop image compression based on clustering algorithm

Abstract

SummaryWith the rapid development of computer software and hardware technology and network technology, aiming at the limitations of the traditional image compression standards and the deficiencies of the existing computer desktop image compression methods. By analyzing the characteristics of computer desktop image, according to the characteristics of desktop compression, a compression scheme based on high efficiency video coding (HEVC) and color clustering is proposed, which divides blocks into text/graphics blocks, natural image blocks and mixed blocks based on the features of histogram information and texture information of blocks. In block division and classification, an adaptive dynamic block classification algorithm is proposed which is different from the traditional block partitioning. Compared with the traditional method, the new block classification algorithm can save the code stream and improve the classification accuracy.