摘要
针对摄像机静止的情况,提出了一种可运用于近景平台的运动目标检测与跟踪方法。采用更新函数对背景进行实时更新,通过差分法检测出运动目标。采用卡尔曼滤波器预测目标参数,在运动目标由近及远与由远及近的情况下,根据预测参数跟踪目标,准确捕获目标轨迹。通过对道路交通场景的测试,算法具有较好的实时性和适应环境变化的能力。
A method of moving object detecting and tracking under stationary camera was presented, which can be used for stationary plat- form. The background was updated by a updating function, with moving objects detected by difference method. With the moving objects from near to far and from far to near,kalman filter can be used for predicting the parameters of objects and tracking it. This algorithm has been test- ed on traffic scenes to show the ability of adapting to environmental change.
出处
《世界科技研究与发展》
CSCD
2013年第1期62-64,共3页
World Sci-Tech R&D
关键词
静止平台
卡尔曼滤波
近景视频
运动目标检测与跟踪
stationary platform
Kalman filtering
close range video
moving object detecting and tracking
