基于纹理特征的“猫眼”效应目标识别方法  

A Recognition Method of “Cat-eye” Effect Target Based on Texture Character

作　　者：李丽[1] 王兴宾[1] 张卫国[2] 

机构地区：[1]北京航空航天大学电子信息工程学院,北京100191 [2]北京信息技术研究所,北京100013

出　　处：《光子学报》2014年第2期131-141,共11页Acta Photonica Sinica

基　　金：航空科学基金项目(No.20070151002)资助

摘　　要：为解决目前方法在动态复杂环境下不能有效区分车灯和瞄准镜及'猫眼'效应目标识别率低的难题,提出了一种基于纹理特征的'猫眼'效应目标识别方法.首先采用列均值相减、指数高通滤波和中值滤波对差分图像进行滤波去除车灯;然后依据'猫眼'效应目标的形状特性进行判别;最后根据'猫眼'效应目标的纹理特征,构造隶属度向量,按照最大隶属度原则,采用模糊综合评判对'猫眼'效应目标进行识别.试验结果表明,与基于形状和频率双重判据的'猫眼'效应目标识别方法相比,该方法有更高的识别率和较低的虚警率.In order to overcome the low recognition rate of ＂ cat eye＂ effect target and ineffective distinction of car lights and telescopic sights, a recognition method of ＂cat-eye＂ effect target based on texture character was proposed. Firstly, car lights were removed by filtering the difference images with the column-mean -subtraction, EHPF and median filter. Secondly, the discrimination was carried out in terms of the features of the shape of the ＂cat-eye＂ effect target. Finally, according to the texture features of the ＂cat-eye＂ effect target, membership vectors were constructed, and the fuzzy comprehensive evaluation was used to recognize the ＂cat-eye＂ effect target under the maximum membership degree principle. Experimental results indicate that the proposed method reduces false alarm rate and improves recognition rate compared with the shape-frequency dual criterions （SFDC） method.

关 键 词：激光成像 目标识别 纹理特征 “猫眼”效应 图像增强 图像配准 图像处理 指数高通滤波 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]