• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2010, Vol. 32 ›› Issue (2): 146-149.doi: 10.3969/j.issn.1007130X.2010.

• 论文 • 上一篇    下一篇

一种基于自适应权值灰关联分析的毫米波/红外融合识别方法

  

  1. (国防科学技术大学ATR重点实验室,湖南 长沙 410073)
  • 收稿日期:2008-08-25 修回日期:2008-11-25 出版日期:2010-01-25 发布日期:2010-01-26
  • 通讯作者: 郭军 E-mail:guojun@nudt.edu.cn
  • 作者简介:郭军(1982), 男, 新疆伊犁人,硕士生,研究方向为图像理解与信息融合;赵和鹏,工程师,研究方向为遥感情报分析;朱长仁,副教授,研究方向为图像理解与信息融合、遥感图像目标识别等。

A Method of MMW/IR Fusion for  Target Recognition Based on Gray  Relational Analysis with Adaptive Weights

  1. (National Key Laboratory of ATR,National University of Defense Technology,Changsha 410073;)
  • Received:2008-08-25 Revised:2008-11-25 Online:2010-01-25 Published:2010-01-26

摘要:

本文提出了一种基于自适应权值灰关联分析和DS理论的毫米波/红外融合目标识别方法。首先对单个传感器提取的特征计算灰关联度作为基本概率赋值,再用DS理论进行融合得到识别结果。在灰关联度计算中,本文提出了使用熵值分析法自适应确定不同特征的权值,提高了系统性能。基于实测数据的实验结果验证了本文的有效性。

关键词: 信息融合, 目标识别, 灰关联分析, 自适应权值, 熵值分析

Abstract:

A method of MMW/IR fusion for target recognition based on gray relational analysis with adaptive weights and the DS theory is presented. Firstly, the gray relational grade of each sensor is computed and treated as the basic probability assignment function. Then, the fusion recognition result is computed by the DS theory. At the step of gray relational grade calculation, an algorithm of entropy analysis is introduced to adaptively determine each weight, and then the system performance is improved. The experimental results based on real data show that the proposed method is effective.

Key words: information fusion;target recognition;gray relational analysis;adaptive weight;entropy analysis

中图分类号: