首页 | 官方网站   微博 | 高级检索  
     

基于图像局部特征的红外小目标检测与跟踪算法
引用本文:成兰,王志杰,姬琇荔.基于图像局部特征的红外小目标检测与跟踪算法[J].红外技术,2008,30(6).
作者姓名:成兰  王志杰  姬琇荔
作者单位:安阳工学院科研处,河南,安阳,455000
摘    要:基于红外序列图像中小目标的典型特征,提出了一种新的红外小目标检测与跟踪算法.利用图像小目标的微分几何特性,计算图像的最小法向曲率,并以此为阈值,获得小目标的候选对象.之后利用多幅图像之间目标的相关性,在时间尺度上基于Kalman滤波算法,实现对目标噪声的过滤并实现目标的跟踪.最后对算法的有效性进行了验证.

关 键 词:目标检测  序列图像  法向曲率

An Novel Algorithm for Infrared Small Targets Detection and Tracking Based on Local Image Features
CHENG LAN,WANG Zhi-Jie,JI Xiu-li.An Novel Algorithm for Infrared Small Targets Detection and Tracking Based on Local Image Features[J].Infrared Technology,2008,30(6).
Authors:CHENG LAN  WANG Zhi-Jie  JI Xiu-li
Affiliation:CHENG LAN,WANG Zhi-Jie,JI Xiu-li(Research office of Anyang Institute of Technology,Henan Anyang 455000,China)
Abstract:Based on the typical feature of small target in sequence infrared images,a new algorithm is proposed for detecting and tracking small targets in sequence images.By using differential geometry property of small object of infrared image,computing minimal normal curvature which is used as the threshold to obtain candidate objects for small target.By using the relationship of objects between different images,blurred objects are filtered by Kalman filter.Finally,the effectiveness of this algorithm is verified.
Keywords:target detection  sequence images  normal curvature  
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-23

京公网安备 11010802026262号