一种压缩比自适应快速矢量量化器的设计与实现  

Design and Implementation for a Compression Ratio Adaptive Algorithm Vector Quantizer

作　　者：王丽娟[1] 余宁梅[1] 王冬芳[1] 马海侠[1] 

机构地区：[1]西安理工大学自动化与信息工程学院,陕西西安710048

出　　处：《西安理工大学学报》2008年第2期215-219,共5页Journal of Xi'an University of Technology

基　　金：国家自然科学基金资助项目(60276017);陕西省自然科学基金资助项目(2006F29);应用材料创新基金资助项目(XA-AM-200714)

摘　　要：设计了一种图像16×16、8×8、4×4动态分块压缩比自适应的快速矢量量化算法。该算法可以根据图像的特征,在较大范围内自适应地变化压缩比,减少重复搜索的运算量,在提升图像压缩比的同时,编码速度也有很大幅度的提高。设计了编码电路,并在FPGA上进行了仿真及验证。结果表明,整个系统最大时钟频率可达76.55 MHz。与其他相关算法相比,该算法的压缩比有很大提高,而平均峰值信噪比降低却很少,可完全满足图像编码的高速和实时性要求。A compression ratio adaptive algorithm for vector quantization （VQ） of a kind of image with 16 × 16, 8 × 8, 4 × 4 dynamic partitioning is designed in this paper. Based on the image features, this algorithm can adaptively change the compression ratio within a very large scope, reduce the computation quantity of repeating searching and improve the encoding speed by a wide margin in simultaneously lifting image compression ratio. Also, the encoding circuit is designed; and the simulation and testing are carried out on FPGA. The results indicate that the maximum clock frequency can reach 76.55 MHz. As compared with some related algorithms, the compression ratio by this algorithm can be greatly improved while the average peak signal to noise ratio （PSNR） can be reduced only a little. However, the high speed and real-time requirements of image encode can be completely satisfied.

关 键 词：矢量量化 压缩比自适应 动态划分 编码电路 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]