摘要
为了在静态和动态场景中均能实现对运动目标的检测与跟踪,提出了基于运动检测和视频跟踪相结合的视频监控方法.建立四参数运动仿射模型来描述全局运动,采用块匹配法对其进行参数估计;采用基于全局运动补偿的Horn-Schunck算法检测出运动目标;使用卡尔曼滤波对运动目标的质心位置、宽度和高度进行跟踪.实验结果表明,该方法能够有效地对静态和动态场景中运动目标进行检测与跟踪.
A visual surveillance system is presented based on the integration of motion detection and visual tracking in static and dynamic scene image sequence.Four parameters model is established for global motion,and the model parameters estimated by block matching.Then moving objects regions were detected by Horn-Schunck algorithm with global motion vectors modified.The center of mass,width and height of moving objects were tracked by Kalman filter.Experimental results showed that this method is effective for moving objects detection and tracking in static and dynamic scenes.
出处
《北京理工大学学报》
EI
CAS
CSCD
北大核心
2009年第10期858-860,876,共4页
Transactions of Beijing Institute of Technology
基金
国家部委预研项目(06104040)
国家重点实验室自主研究课题(ZDKT08-05)
关键词
动态场景
全局运动
光流法
卡尔曼滤波器
dynamic scene
global motion
optical flow method
Kalman filter