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

基于时空特性的运动目标检测算法研究
引用本文:张明艳,许钢,孟樱.基于时空特性的运动目标检测算法研究[J].安徽工程科技学院学报,2013(4):58-61.
作者姓名:张明艳  许钢  孟樱
作者单位:安徽工程大学电气工程学院,安徽芜湖241000
基金项目:安徽工程大学青年基金资助项目(20LOYQ036);安徽省教育厅产学研重点基金资助项目(KJ2011A035)
摘    要:采用时序分割虽然能够检测到运动目标区域,却不能准确地提取出目标轮廓.本文提出了基于时空特性的运动目标检测算法.由背景减法检测到运动目标的位置、形状、大小信息,然后采用以彩色梯度为外部能量的活动轮廓(snake)算法获得准确的运动对象轮廓.实验证明,本算法能够很好地提取出运动目标.

关 键 词:目标检测  时空特性  彩色梯度

Moving object detection algorithm based on time-spatial properties
ZHANG Ming-yan,XU Gang,MENG Ying.Moving object detection algorithm based on time-spatial properties[J].Journal of Anhui University of Technology and Science,2013(4):58-61.
Authors:ZHANG Ming-yan  XU Gang  MENG Ying
Institution:(Coll. fo Elec. Engn. ,Anhui Polytechnic University, Wuhu 241000, China)
Abstract:A new algorithm based on time-spatial properties was displayed in this paper. It is known to all that background subtraction can detect the position, shape and size of moving objects,but it can't show the exact contour of moving objects. So background subtraction is employed to find the field of objects and the accurate objects" contours are extracted by active contour (snake) using color gradient as its ex- ternal energy. And the experiment shows that this method has a good result in moving object detection.
Keywords:object detection  time-spatial properties ~ color gradient
本文献已被 维普 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