首页 | 本学科首页   官方微博 | 高级检索  
     

改进的支持向量机低分辨雷达目标分类算法
引用本文:陈志仁,顾红,苏卫民,王钊. 改进的支持向量机低分辨雷达目标分类算法[J]. 系统工程与电子技术, 2017, 39(11): 2456-2462. DOI: 10.3969/j.issn.1001-506X.2017.11.10
作者姓名:陈志仁  顾红  苏卫民  王钊
作者单位:1. 南京理工大学电子工程与光电技术学院, 江苏 南京 210094;;2. 中国电子科技集团公司第五十四研究所, 河北 石家庄 050081
摘    要:针对雷达目标识别系统中存在库外目标、各特征属性分类性能不一致,低分辨雷达不同类别目标特征值分布区间混叠的问题。文中基于支持向量机分类算法,对目标样本特征值进行置信度分析,获得置信区间,利用代价函数确定拒识门限,对库外目标以及混叠区域目标进行拒识分析;并根据训练样本先验识别率使用注水原理分配特征属性权重;在此基础上利用支持向量机进行目标识别。雷达实测数据实验结果表明,引入置信拒识分析与注水分配加权后,分类器性能得到了有效改善。


Improved support vector machine target classification algorithm for low-resolution radar
CHEN Zhiren,GU Hong,SU Weimin,WANG Zhao. Improved support vector machine target classification algorithm for low-resolution radar[J]. System Engineering and Electronics, 2017, 39(11): 2456-2462. DOI: 10.3969/j.issn.1001-506X.2017.11.10
Authors:CHEN Zhiren  GU Hong  SU Weimin  WANG Zhao
Affiliation:1. School of Electronics Engineering & Optoelectronic Technology, Nanjing University of Science &Technology, Nanjing 210094, China;2. The 54th Research Institute of CECT, Shijiazhuang 050081, China
Abstract:In order to solve the problems of the targets being outside of the database, inconsistent classification of each feature in the radar automatic recognition system and aliasing of different types of targets in the low resolution radar, based on the support vector machine classification algorithm, the confidence interval is obtained by analyzing the sample characteristic values, the cost function is used to determine the rejection threshold, and it is used to reject outside targets and aliasing targets. According to the recognition rate of training samples, the water-filling theory is introduced to the attribute weights assignment. Then, the support vector machine is used for target recognition. The experiments with the radar measured data show that the performance of the support vector machine has been well improved.
Keywords:
点击此处可从《系统工程与电子技术》浏览原始摘要信息
点击此处可从《系统工程与电子技术》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号