摘要
提出了一种模糊聚类和模糊模式识别相结合的目标识别方法,并成功应用于海上舰船识别分类;同时引入聚类分析有效性评价的F统计量,实现了模糊聚类的自适应性,避免了聚类数目选取上存在的主观性。对于给定特征的海上舰船目标,仿真实现了对目标的聚类分析,获得目标的分类并形成标准模型库,并通过模糊模式识别对后继获得的目标特征样本在标准模型库中进行匹配,应用最大贴近度原则完成目标识别。仿真结果表明:对于复杂的战场环境,两种方法的结合是可行和有效的,可以满足战时实时性和准确性的要求,具有一定的应用前景。
A target recognition method that combines fuzzy clustering with fuzzy pattern recognition is presented, which has been successfully applied to identify vessels on the sea. F- statistic is introduced to evaluate the effectiveness of the cluster analysis for realizing a self- adaptive fuzzy clustering and avoiding subjectivity in the selection of cluster number. For the targets (vessels) on the sea with given characteristics, the target clustering analysis is accomplished by simulation, and a classification standard model database is formed. For the characteristics of the target obtained subsequently, fuzzy pattern recognition is used for matching them in the standard model database. Then maximum Proximity Degree principle is used for implementing the target recognition. Simulation results showed that in the complex battlefield environment, the combination of the two methods is feasible, effective and can satisfy the real - time and accuracy requirements.
出处
《电光与控制》
北大核心
2007年第4期35-38,共4页
Electronics Optics & Control
基金
船舶工业国防科技预研基金项目(03J3.6.1)
关键词
模糊聚类分析
模糊模式识别
目标分类
目标识别
贴近度
fuzzy clustering analysis
fuzzy pattern recognition
target classification
target recognition
proximity degree