摘要
针对传统粒子算法容易陷入局部极值而产生过早收敛的问题,提出一种基于遗传粒子优化算法。利用遗传算法在全局搜索方面的优势,对粒子算法进行改进,从而得到一种遗传粒子优化算法。在图像的分类中,传统K-Means聚类算法聚类中心的选择影响较大的问题,引入遗传粒子优化算法,对聚类中心进行优化,从而避免聚类中心的随机选择给图像分类精度带来很大的影响。最后,通过系统仿真比较,验证了该算法的优势。
Based on the traditional PSO algorithm is easy to fall into local extremum and premature convergence,the paper proposes a genetic algorithm based on particle swarm optimization. The advantage of global search using genetic algorithm,the particle algorithm is improved,resulting in a genetic particle swarm optimization algorithm. In image classification,the traditional K- means clustering algorithm clustering center selection effect is a big problem that is introduced into the genetic particle swarm optimization algorithm,to optimize the clustering center,thus avoiding the random selection of cluster center bring great influence to the accuracy of image classification. Finally,through the system simulation,verified the advantages of the proposed algorithm.
出处
《自动化与仪器仪表》
2016年第7期163-164,共2页
Automation & Instrumentation
关键词
粒子算法
遗传算法
图像分类
K-MEANS聚类
仿真
particle algorithm
genetic algorithm
image classification
K-Means clustering
simulation
