首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于检测识别的实孔径雷达景象匹配定位方法
引用本文:杨卫东,左峥嵘,张天序.基于检测识别的实孔径雷达景象匹配定位方法[J].华中科技大学学报(自然科学版),2005,33(2):25-27.
作者姓名:杨卫东  左峥嵘  张天序
作者单位:1. 华中科技大学,图像识别与人工智能研究所,湖北,武汉,430074
2. 图像信息处理与智能控制教育部重点实验室,湖北,武汉,430074
基金项目:国家自然科学基金重点项目(60135020).
摘    要:针对恶劣成像条件下的机场目标实孔径雷达图像匹配问题,提出了一种基于目标检测识别的定位方法,通过线特征滤波预处理和投影变换检测机场跑道线,进一步用模板匹配识别跑道端点以进行定位.对实际图像试验结果表明该算法能够准确定位灰度相关算法难以正确匹配的强噪声、低对比度实时图像,并且具有参考图制备简单,实时性强的特点.

关 键 词:实孔径雷达  目标识别  景象匹配  机场检测
文章编号:1671-4512(2005)02-0025-03
修稿时间:2004年9月24日

A method of registration for real aperture radar scene based on airfield detection and recognition
Yang Weidong,Zuo Zhengrong,Zhang Tianxu.A method of registration for real aperture radar scene based on airfield detection and recognition[J].JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY.NATURE SCIENCE,2005,33(2):25-27.
Authors:Yang Weidong  Zuo Zhengrong  Zhang Tianxu
Institution:Yang Weidong Zuo Zhengrong Zhang Tianxu Doctoral Candidate, Institue for Pattern Recognition & AI.,Huazhong Univ. of Sci. & Tech.,Wuhan 430074,China.
Abstract:A method based on airfield detection and recognition was proposed, aiming to the registration for airfield under ill condition of real aperture radar imaging. Through the linear feature enhancement and Randon transform to locate the runway line, then using model matching to find the exactly position of the runway end for registration of the airfield. The experiment showed that this method could accurately registration the airfield, even for serious speckle and low contrast radar image. It has the benefit of simplified reference map preparation and less computation than the gray correlation algorithm.
Keywords:real aperture radar  target recognition  scene matching  airfield detection
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