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

红外序列图像目标的高效跟踪
引用本文:胡光,袁胜智.红外序列图像目标的高效跟踪[J].烟台大学学报(自然科学与工程版),2011,24(2).
作者姓名:胡光  袁胜智
作者单位:1. 烟台大学计算机科学与技术学院,山东烟台,264005
2. 海军航空工程学院兵器科学与技术系,山东烟台,264000
摘    要:针对Mean shift算法中仅使用灰度特征难于实现对红外序列图像目标准确跟踪的问题,提出了利用边缘-灰度特征相融合的目标跟踪算法.为了实现红外序列图像中目标的高效跟踪,在研究Mean shift灰度特征的基础上,根据红外序列图像的边缘方向信息与灰度信息构建了边缘-灰度目标模型,将此模型应用到目标跟踪算法中,提高了Mean shift算法在红外序列图像目标跟踪过程中的运算能力.实验证明,该方法能够增加灰度直方图对目标模型的描述能力,提高了目标跟踪算法的鲁棒性.

关 键 词:灰度特征  相似度函数  边缘特征  Mean  shift算法  红外序列图像

Efficient Target Tracing in Infrared Sequence Images
HU Guang,YUAN Sheng-zhi.Efficient Target Tracing in Infrared Sequence Images[J].Journal of Yantai University(Natural Science and Engineering edirion),2011,24(2).
Authors:HU Guang  YUAN Sheng-zhi
Institution:HU Guang1,YUAN Sheng-zhi2(1.School of Computer Science and Technology,Yantai University,Yantai 264005,China,2.Department of Science and Tech-nology of Weapons,Naval Aeronautical Engineering Institute,Yantai 264000,China)
Abstract:In order to trace target in infrared sequence images efficiently,a target tracing algorithm is proposedby edge-gray features fusion and the edge-gray target model is established according to edge and gray informa-tion in infrared sequence images by studying gray features on Mean shift algorithm.The model can improvecomputing ability of Mean shift algorithm in the process of tracing target in infrared sequence images.Also,itis shown that our approach can help to improve the ability of gray histogram in descr...
Keywords:gray feature  proximity functions  edge feature  Mean shift algorithm  infrared sequence images  
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