采用粒子群优化粒子滤波的红外目标提取算法
投稿时间:2009-12-28  最后修改时间:2009-12-28  点此下载全文
引用本文:周越,毛晓楠.采用粒子群优化粒子滤波的红外目标提取算法[J].红外与毫米波学报,2010,29(1):63~68].ZHOU Yue,MAO Xiao-Nan.INFRARED TARGET EXTRACTION ALGORITHM BY USING PARTICLE SWARM OPTIMIZATION PARTICLE FILTER[J].J.Infrared Millim.Waves,2010,29(1):63~68.]
摘要点击次数: 5596
全文下载次数: 6175
作者单位E-mail
周越 上海交通大学 meureka@126.com 
毛晓楠 上海航天控制工程研究所  
基金项目:国家自然科学基金(60772097)和航空科学基金(2008ZC57)
中文摘要:提出了一种新的基于粒子群优化粒子滤波(PSOPF)的红外目标提取算法,将红外目标提取阈值的计算问题看作系统状态估计问题.在粒子滤波的框架下,建立了关于灰度—方差加权信息熵和像素点灰度值的阈值状态空间,建立了基于粒子群优化算法思想的系统状态转移模型,建立了基于红外目标提取效果评价函数的系统观测模型,它有效综合了红外图像中灰度、信息熵、梯度、像素点的空间位置等信息.最后,以粒子的加权平均估计目标提取的阈值.实验结果表明,该方法是有效且稳健的.
中文关键词:粒子滤波  粒子群优化  目标提取  灰度—方差加权信息熵  
 
INFRARED TARGET EXTRACTION ALGORITHM BY USING PARTICLE SWARM OPTIMIZATION PARTICLE FILTER
Abstract:A novel infrared target extraction algorithm based on particle swarm optimization particle filter(PSOPF) was proposed. The problem of infrared target extraction was analyzed and solved in the view of state estimation. In the framework of particle filter, the threshold state space on the gray-variance weighted information entropy and the grey value of each pixel was based on extraction results evaluation function, which integrated grey, entropy, gradient and spatial distribution of pixels. Finally, the weighted average of all the particles was used as target extraction threshold. The experiment results prove that the proposed algorithm is effective and robust.
keywords:particle filter  particle swarm optimization  target extraction  gray-variance weighted information entropy  
查看全文  查看/发表评论  下载PDF阅读器

版权所有:《红外与毫米波学报》编辑部