动态模糊矢量量化算法

由于传统的Ｋ－均值算法在用于矢量量化时强烈依赖初始码书的选取，如果初始码书选取不好，则很容易陷入局部最小点；而Ｂｅｚｄｅｋ的模糊Ｋ－均值算法由于计算量很大，也很少用于矢量量化的设计码书，因此人们一直在寻找收敛速度和收敛效果两者性能较好的算法，在研究Ｎｉｃｏｌａｏｓ等人提出的模糊矢量量化（ＦＶＱ）算法基础上，针对ＦＶＱ算法收敛过程存在的总理２，并从收敛结构和收敛策略出发，提出了一种动态的法在收敛速度

Dynamic Fuzzy Vector Quantization Algorithm

K-means algorithm applied invector quantization strongly dep ends on the selection of the initial codebook, and if notgiven a good initial c odebook it can easily be trapped in local minima. Furthermore,Bezdek's fuzzy K -means algorithms are computationally expensive so that they areimpractical in codebook design. So, people have been researching those algorithms whichcan ac hieve good performance in the convergent speed of algorithms and the quality of thereconstructed image. Analyzing the fuzzy vector quantization algorithm (FVQ) presented byNicolaos.B.K. and aiming at the irrational convergent procedure of the algorithm, from theaspect of convergent structure and strategy the paper p resents a dynamic fuzzy vectorquantization algorithm (DFVQ) and gives two concr ete methods based on the idea of thepresented algorithm. Experiments show the p resented methods markedly accelerate theconvergent procedure and improve the qu ality of convergence.