首页 | 本学科首页   官方微博 | 高级检索  
     

基于辅助粒子滤波的红外小目标检测前跟踪算法
引用本文:胡洪涛,敬忠良,胡士强.基于辅助粒子滤波的红外小目标检测前跟踪算法[J].控制与决策,2005,20(11):1208-1211.
作者姓名:胡洪涛  敬忠良  胡士强
作者单位:上海交通大学,电子信息与电气工程学院,上海,200030
基金项目:国家自然科学基金项目(60375008),高校博士点基金项目(20020248),上海市科技攻关重大预研项目(035115009)
摘    要:研究低信噪比复杂环境下的红外小目标检测和跟踪问题,提出了基于辅助粒子滤波的检测前跟踪算法.首先使用形态学滤波算法对图像进行白化预处理;然后在跟踪阶段采用辅助粒子滤波算法估计目标运动状态,在检测阶段利用跟踪滤波器的输出构造似然比,并进行似然比检验.对真实红外图像序列的实验表明,该算法可成功跟踪和检测信噪比为2的小目标,且其性能优于传统的检测前跟踪算法.

关 键 词:检测前跟踪  辅助粒子滤波  红外图像序列  小目标
文章编号:1001-0920(2005)11-1208-04
收稿时间:2004-11-15
修稿时间:2005-04-19

Auxiliary Particle Filter Solution to Track-before-detect for Small Infrared Targets
HU Hong-tao,JING Zhong-liang,HU Shi-qiang.Auxiliary Particle Filter Solution to Track-before-detect for Small Infrared Targets[J].Control and Decision,2005,20(11):1208-1211.
Authors:HU Hong-tao  JING Zhong-liang  HU Shi-qiang
Affiliation:School of Electrical and Information Engineering, Shanghai Jiaotong University, Shanghai 200030, China
Abstract:The problem of detecting and tracking dim small targets in a sequence of infrared images with very low SNR and complex background is investigated.A track-before-detect algorithm for small target is developed based on auxiliary particle filter.In the proposed algorithm,firstly,morphological filtering is performed on image for prewhitening.Then,the auxiliary particle filter is used to estimate the state of the target in track stage.Finally,some values of the filter's output are used to approximately construct the likelihood ratio for hypothesis test in detection stage.Experimental results with the real image sequences verify that the proposed algorithm can successfully detect and track small target with SNR equaling to about 2.0,and the performance is better than traditional track-before-detect algorithm.
Keywords:Track before detect  Auxiliary particle filter  Infrared image sequences  Small target
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《控制与决策》浏览原始摘要信息
点击此处可从《控制与决策》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