中文摘要：

针对标准猫群算法在矢量量化码书设计中收敛速度慢及易陷入局部最优的缺点,将标准猫群优化算法和云模型相结合,提出了一种基于云模型猫群算法.通过运用云发生器建立猫个体变异程度和适应值大小的 关系,实现猫群搜索的自适应调节,从而增强种群多样性、提高收敛速度,避免局部最优.仿真实验证明,改进的 算法较其他同类型算法在收敛性、类间离散度和矢量量化不均匀度等方面有较大的提升.

英文摘要：

Standard cat swarm optimization algorithm has slow convergence rate and easily fall into local optimum in vector quantization codebook design. Therefore, a new codebook design algorithm is proposed combining standard cat swarm optimization algorithm and cloud model. By using cloud generator to set up the relationship between individual variation of the cat and the value of fitness, search range of cats will realize adaptive control, improving the diversity of populations and the convergence rate and avoiding local optimum. Experiment shows that the improved algorithm has higher performance than other similar algorithms in convergence, class scatter and vector quantization uniformity.