首页 | 本学科首页   官方微博 | 高级检索  
     

一种基于红外图像的目标自动识别算法
引用本文:李炯,雷虎民. 一种基于红外图像的目标自动识别算法[J]. 航空计算技术, 2005, 35(4): 26-28
作者姓名:李炯  雷虎民
作者单位:空军工程大学,导弹学院,陕西,三原,713800
摘    要:红外成像制导具有在各种复杂战术环境下自主搜索、捕获、识别和跟踪目标的能力,代表了当代红外制导技术的发展趋势。提出了一种红外图像预处理、跟踪、分类的自动目标识别算法,利用小波变换、形态学方法对红外图像进行预处理,提取不同频带的惯性不变矩作为特征量,利用神经网络进行分类识别,结果表明该算法具有很高的识别率,对于精确制导武器的目标识别研究具有一定的参考价值。

关 键 词:红外图像  小波变换  特征提取  神经网络  自动目标识别
文章编号:1671-654X(2005)04-0026-03
修稿时间:2005-05-26

A Method of Automated Recognition and Classification based on Infrared Images
LI Jiong,LEI Hu-min. A Method of Automated Recognition and Classification based on Infrared Images[J]. Aeronautical Computer Technique, 2005, 35(4): 26-28
Authors:LI Jiong  LEI Hu-min
Abstract:Infrared imaging homing is development trend of infrared homing technology,which has independent capability in target recognition and tracking.A novel ATR algorithm including image pre-processing,tracking and classification for sequence of infrared(IR) targets is presented in this paper.Decompose the infrared image by wavelet transform,and then extract the moments as the features to be recognized by BP neural network.The experimental result indicates that this method has very high discernment rate and definite practical value to target recognition study of accurate guidance weapon.
Keywords:infrared image  wave transform  feature extraction  neural network  automatic target recognition
本文献已被 CNKI 维普 万方数据 等数据库收录!
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