基于模糊聚类和卡尔曼滤波的运动目标检测

Moving objects detection based on fuzzy clustering and Kalman filtering

提出了一种基于模糊聚类和卡尔曼滤波的多运动目标检测的技术,并将其应用于车辆的检测与跟踪。文中采用了改进的模糊C均值聚类算法,对隶属度矩阵进行了修正,加快了聚类的收敛速度;然后,运用了卡尔曼滤波对运动目标进行运动轨迹跟踪,根据视频序列的实际情况构造了相应的状态方程以及增益矩阵,对多个运动目标同时进行独立跟踪,减少了目标搜索的盲目性,提高了跟踪的可靠性和效率。

A kind of technique for detection of multiple moving objects based on fussy clustering and Kalman filtering was brought forward, and has been applied to vehicle detection and tracking. An improved fussy C mean clustering algorithm was used, in which the matrix of grade of membership was modified in order to speed up convergent velocity. Kalman filtering was used to track moving target. Corresponding state equation and plus matrix were constructed based on video sequence to track multiple moving objects at the same time. It can achieve fine object searching with more reliability and efficiency.