|
|
基于特征值的激光成像雷达目标识别算法
Target recognition algorithm of laser imaging radar based on feature value
下载全文
基于特征值的激光成像雷达目标识别算法.pdf
(844 KB, 下载次数: 1)
作 者:李春苗[1] 高剑波[2] 杨镇源[2] 周昕[1] 张华[2] 冷杰[2] 宋昭[2]
LI Chun-miao,GAO Jian-bo,YANG Zhen-yuan,ZHOU-Xin,ZHANG Hua,LENG Jie,SONG Zhao( 1. Institute of Electronic Information, Sichuan University, Chengdu 610065, China ; 2. Southwest Iustitute of Technical Physics,Chengdu 610041 ,China)
机构地区:[1]四川大学电子信息学院,四川成都610065 [2]西南技术物理研究所,四川成都610041
出 处:《激光与红外》 CSCD 2015年第45卷第3期 335-338页,共4页
Laser & Infrared
摘 要:提出了一种基于特征值的激光成像雷达目标识别算法。首先从激光成像雷达中获取目标场景的距离像,将场景中各个目标分离,分别提取各个目标距离像的特征值,然后用网格搜索算法对支持向量机参数进行优化,最后应用支持向量机对分离出来的三种目标进行分类识别。仿真实验表明:该方法简单易于实现,识别效果较好。
Target recognition algorithm of laser imaging radar based on feature value was put forward. Firstly range images of the object scene from laser imaging radar were achieved,and each object in the scene was separated,and the feature values were extracted from range images. Secondly the grid search algorithm was used to optimize parameters of support vector machine( SVM). At last,SVM was used to classify and recognize the segregated targets. Simulation experiment shows that the method is simple and easy to implement,and has good recognition effect.
关键词:特征值 支持向量机 激光雷达 目标识别
feature value ; support vector machine ; laser imaging radar ; target recognition ;
|
|