【作者】 罗雪晖； 李霞； 张基宏；

【Author】 Luo Xuehui Li Xia Zhang Jihong (College of Information Engineering,Shenzhen University,Shenzhen 518060)

【机构】 深圳大学信息工程学院；

【摘要】 本文提出一种基于自适应蚁群优化的矢量量化码书设计算法。该算法利用人工蚁群系统中蚂蚁通过信息量留存寻找最优路径的机制,合理设计转移概率、禁忌列表、信息量更新方式及自适应地调整截取转移概率的参数。实验结果表明,该算法能获得质量较好的码书和主观效果较好的译码图像。更多还原

【Abstract】 A VQ codebook design algorithm based on the adaptive ant colony algorithm is proposed.Inspired by the ability of real ants to find the shortest path,we proposed the ant clustering based codebook design algorithm with emphasis on the appropriate design of the transition probability,the tabu list as well as the way pheromone was updated,in particular the adaptive adjustment of the intercept parameter of the transition probability.Experimental results show that the algorithm can attain a codebook with much improved PSNR and the decode image has better subjective performance.更多还原