现代电子技术2016，Vol.39Issue(19)：169-173,178,6.DOI:10.16652/j.issn.1004-373x.2016.19.041
基于局部不变特征与Camshift复杂环境跟踪技术研究
Research on complex environment vehicle tracking technology based on local invariant feature and Camshift
摘要
Abstract
Since the traditional tracking methods are failed due to the complex environments of vehicle deformation,haze weather,background noise,illumination variation and occlusion interference existing in urban road traffic,the vehicle target tracking new algorithm of urban road traffic is proposed,which integrates HSV color model and local invariant feature with the continuously adaptive mean shift(Camshift)algorithm. The multi⁃feature template based on color and gray level is established. The sample principal component analysis method of statistic analysis is used to reduce the feature dimension to improve the effi⁃ciency of feature matching. The spatial distribution of weighted features is defined and calculated,so the algorithm can better deal with the complex traffic environment,and ensure the stable and accurate tracking. The experimental results show that the new algorithm has high recognition rate,good real⁃time performance,and strong robust to resist environmental interference.关键词
特征融合/Camshift/SIFT算法/降维/目标跟踪Key words
feature fusion/Camshift/SIFT algorithm/dimension reduction/target tracking分类
信息技术与安全科学引用本文复制引用
厉丹,鲍蓉,党向盈,程德强..基于局部不变特征与Camshift复杂环境跟踪技术研究[J].现代电子技术,2016,39(19):169-173,178,6.基金项目
住房城乡建设部科学技术计划项目（2014-K5-027）；江苏省高校自然科学研究面上项目（15KJB520033）；国家自然科学基金资助项目（51204175，U1261105）；江苏省大型工程装备检测与控制重点建设实验室开放课题（JSKLEDC201224）；徐州工程学院重点培育项目 （）
