摘要
针对传统的聚类算法在聚类划分问题上还存在着划分效果能力较差等问题,在确保GA算法全局性能和收敛速度的前提下,设计了一种与进化代数相关联的交叉概率和与个体适应度相关联的变异概率。将该算法应用到图像聚类,实验仿真结果表明,对比于k均值聚类算法,该算法具有较好的聚类划分效果。
According to the traditional clustering algorithm in clustering partition problems still exist and the partition effect ability is poor.Based on the overall performance and ensure the GA algorithm convergence speed,under the premise of designing an evolutionary associated with individual fitness crossover probability and mutation probability associated.The algorithm is applied to image clustering,the experimental simulation results show that,compared with the k-means clustering algorithm,the proposed algorithm has good effect of cluster division.
出处
《信息技术》
2011年第4期190-192,196,共4页
Information Technology
关键词
遗传算法
聚类分析
交叉概率
变异概率
genetic algorithm
clustering analysis
crossover probability
mutation probability
