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

多特征的UAV快速目标识别算法仿真
引用本文:李大健,刘慧霞,张清,席庆彪,郭海帆.多特征的UAV快速目标识别算法仿真[J].火力与指挥控制,2012,37(3):35-38.
作者姓名:李大健  刘慧霞  张清  席庆彪  郭海帆
作者单位:1. 西北工业大学第365研究所,西安,710065
2. 西北工业大学自动化学院,西安,710072
3. 西北工业大学第365研究所,西安710065;西北工业大学自动化学院,西安710072
4. 西南电子设备研究所,成都,610036
基金项目:国家自然科学基金资助项目(61074155)
摘    要:针对UAV(Unmanned Aerial Vehicle)侦察图像快速目标识别问题,重点展开基于多特征的UAV快速目标识别算法的仿真研究。算法结合图像的不变矩特征和SIFT特征,首先用不变矩特征构造适应度函数并利用遗传算法的全局搜索能力,在侦察图像中进行搜索,快速提取出可能包含目标的感兴趣区域(ROI,Region of Interest);然后采用尺度不变特征变换算法(SIFT,Scale Invariant Feature Transform)在ROI区域中进行匹配识别,从而确定目标的精确位置。仿真结果表明:算法具有较强的鲁棒性,能有效地识别飞机目标并显著减少识别时间,为UAV系统提供了一种近实时的目标识别方法。

关 键 词:无人机  目标识别  不变矩  遗传算法  SIFT

Fast UAV Target Recognition Algorithm Based on Multiple Features
LI Da-jian , LIU Hui-xia , ZHANG Qing , XI Qing-biao , GUO Hai-fan.Fast UAV Target Recognition Algorithm Based on Multiple Features[J].Fire Control & Command Control,2012,37(3):35-38.
Authors:LI Da-jian  LIU Hui-xia  ZHANG Qing  XI Qing-biao  GUO Hai-fan
Institution:1.No.365 Research Institute,Northwest Polytechnical University,Xi’an 710065,China,2.School of Automation,Northwest Polytechnical University,Xi’an 710072,China,3.Southwest China Research Institute of Electronic Equipmet,Chengdu 610036,China)
Abstract:To settle the problem of identifing the target in the UAV(Unmanned Aerial Vehicle) reconnaissance image in real time,a UAV fast target identify algorithm is proposed,which is based on invariant moments and the SIFT(Scale Invariant Feature Transform) features of the image.Firstly,similarity function is designed with invariant moments,and a fitness function of GA(genetic algorithm) is taken to search the image globally,the region of interest(ROI),which may contain the target,can be extracted quickly.Secondly the location of the object can be determined accurately by SIFT algorithm in the ROI region.Simulation results show that the method is of strong robustness and can effectively identify the plane targets in airport with the ability to reduce the recognition time significantly.
Keywords:UAV  target recognition  invariant moments  genetic algorithm  SIFT
本文献已被 CNKI 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