基于粒子群K均值聚类的空中目标识别  

Air targets recognition based on particle swarm optimization and K-means clustering

作　　者：王少蕾[1] 陈维义[1] 

机构地区：[1]海军工程大学兵器工程系,湖北武汉430033

出　　处：《舰船科学技术》2012年第12期30-34,共5页Ship Science and Technology

摘　　要：舰艇编队面临的空中威胁日益复杂多样,正确、快速地识别目标是赢得对空防御作战的前提。在分析各传感器测量得到的空中目标威胁特征指标的基础上,将舰艇编队防空作战目标识别问题转化为最优聚类问题,建立基于粒子群聚类的目标类型识别模型。通过主成分分析将样本各特征值标准化、降维投影到新的特征空间,引入粒子群优化算法构建最优聚类识别模型,实例分析表明该方法有效,计算速度快,降低了实用的复杂度,提高了目标识别的可靠性。The threat which the formation of ships will face is much more complex and various than ever before. How to recognise targets right and fast is the precondition of air defence. On the basis of analysing the information from surveillance sensors,the targets recognition of warship formations＇ air defence is translated into optimal clustering problems, and the model for identifying air-targets is established. Through standardizing the attribute values of samples, the principal components analysis （PCA）and particle swarm optimization（ PSO ）are introduced and the best solution is established. The model can solve such problems quickly and correctly, and help the commanders make decisions effectively. The analysis of example shows that the algorithm is effective, the computing rate is fast, the reliability of recognition is improved and the complexity of algorithms applied is reduced.

关 键 词：目标识别 主成分分析 粒子群优化 K均值聚类 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]