论文部分内容阅读
针对粒子群优化算法(PSO)应用于矢量量化时,最优粒子对与其对应维度距离较大的粒子缺乏有效指导问题,提出适用于矢量量化的改进粒子群优化算法(IPSO_VQ).该算法通过建立粒子与榜样粒子的维度映射关系,以基于映射关系的维度学习代替对应维度学习关系,使粒子相关维度间的学习有一定相关性,增强算法局部搜索能力.同时,借鉴广泛学习粒子群优化(CLPSO)算法中的广泛学习思想,并将其应用于基本 PSO 中的全局最优位置学习部分,通过对多个粒子的广泛学习,增加种群的多样性.实验结果表明该算法有效避免种群早熟收敛,从而使解码恢复图像获得更高的主客观质量.
In order to solve the problem of particle size optimization, an optimal particle swarm optimization algorithm (IPSO_VQ) is proposed for vector quantization based on particle swarm optimization (PSO) Particles and example particles, the dimension learning based on mapping relationship is replaced by the dimension learning relationship based on the mapping relationship so that the correlation between the relevant dimensions of the particles is related to enhance the local search ability of the algorithm.At the same time, this paper draws lessons from the extensive study of particle swarm optimization (CLPSO ) Algorithm and apply it to the global optimal position learning part in the basic PSO to increase the diversity of the population through the extensive learning of multiple particles.The experimental results show that this algorithm can effectively avoid the premature convergence of the population, So that the decoded image restoration to obtain a higher subjective and objective quality.