一种基于背景模型的运动目标检测与跟踪算法

本文提出了一种静止摄像机条件下的运动目标检测与跟踪算法．它以一种改进的自适应混合 高斯模型为背景更新方法，用连通区检测算法分割出前景目标，以Kalman滤波为运动模型实 现对运动目标的连续跟踪．在目标跟踪时，该算法针对目标遮挡引起的各种可能情况进行了 分析，引入了对运动目标的可靠性度量，增强了目标跟踪的稳定性和可靠性．在对多个室外 视频序列的实验中，该算法显示了良好的性能，说明它对于各种外部因素的影响，如光照变 化、阴影、目标遮挡等，具有很强的适应能力．

MOVING OBJECT DETECTION AND TRACKING BASED ON BACKGROUND SUBTRACTION

An approach to detecting and tracking moving objects with a static camera is presented in this paper. A modified mixture Gaussian model is used as the adaptive background updating method. Foreground objects are segmented based on an improved binary connected component analysis. Kalman filtering is used for object tracking. To deal with the problems of occlusion between objects in tracking, various situations are analyzed and a measure of reliability of moving objects is adopted which makes the tracker more effective. Experiments on several outdoor video streams that show convictive object detection and tracking performance demonstrate its strong adaptability to lighting changes, shadows and occlusions.