| 一种缺损图像的军事目标识别方法 |
| |
| 引用本文： | 蒋少华,王乘,陈雪菘,朱洪波.一种缺损图像的军事目标识别方法[J].小型微型计算机系统,2010,31(6). |
| |
| 作者姓名： | 蒋少华 王乘 陈雪菘 朱洪波 |
| |
| 作者单位： | 1. 华中科技大学,数字化工程中心,湖北,武汉,430074;湖南师范大学,计算机部,湖南,长沙,410081
2. 华中科技大学,数字化工程中心,湖北,武汉,430074 |
| |
| 摘 要： | 在军事目标识别领域,多源图像融合可以消除军事伪装和遮挡的影响,但还需要借助"先局部,再整体"的识别方式来提高识别准确度.本文回顾了目标识别的研究现状,鉴于拐点特征是旋转、平移和缩放不变量,给出适用于多源融合二值缺损图像的军事目标识别的边缘拐点特征的定义及几种阈值的选取方法、拐点特征模板的组织、以及局部匹配度和整体匹配度的计算公式.对一些缺损枪支图片的实验结果表明,本文所介绍的方法具有良好的效果.
|
| 关 键 词： | 模式识别 拐点 缺损图像 军事伪装 多源融合 军事目标 |
Method of Recognizing Military Object from Occluded Images |
| |
| JIANG Shao-hua,WANG Cheng,CHEN Xue-song,ZHU Hong-bo.Method of Recognizing Military Object from Occluded Images[J].Mini-micro Systems,2010,31(6). |
| |
| Authors: | JIANG Shao-hua WANG Cheng CHEN Xue-song ZHU Hong-bo |
| |
| Affiliation: | JIANG Shao-hua1,2,WANG Cheng1,CHEN Xue-song1,ZHU Hong-bo11(Digital Engineer Center,Huazhong University of Science , Technology,Wuhan 430074,China)2(The Computer Education Department of Hunan Normal University,Changsha 410081,China) |
| |
| Abstract: | For military object recognition,it needs not only multi-source fusion technology but also from local to global to eliminate the impact of military camouflage from occluded images.First,studying situation is presented.Second,we focuses on the issue of definition of corner feature vector,which is invariant to shift,rotation,and scale,Then,the threshold value definition,composition of corner feature templet,local goodness of fit(LGOF)and global goodness of fit(GGOF)are discussed.All above are well suitable to ... |
| |
| Keywords: | pattern recognition corner occluded image military camouflage multi-source fusion military object |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
