固定场景下的运动检测与运动跟踪

Moving object detect and track in stable scene

提出了一种检测运动物体,跟踪运动物体的方法。用混合高斯模型得到运动人体的区域,通过卡尔曼滤波对人体进行跟踪,并利用人体的颜色信息进行识别。该方法能够较好解决应用于室外的视频监视系统中的光照问题,具有较快计算速度,满足实时系统的要求。

An approach to detecting and tracking a moving object is presented. Moving areas about human are segmented by using hybrid Gaussian model as background, tracked by Kalman filter, and recognized by using a color-based model. This method can do a great deal with illumination problem about video surveillance system using out-door, and can compute fast enough to suffice realtime system.