基于移动监控装置的实时运动目标跟踪算法  

Real-time object tracking algorithm for mobile surveillance systems

作　　者：齐建冬 徐卫军[1] 董宝良[1] 

机构地区：[1]华北计算技术研究所指挥控制系统部,北京100083

出　　处：《计算机工程与设计》2014年第9期3234-3237,3294,共5页Computer Engineering and Design

基　　金：国家863高技术研究发展计划基金项目(2010AA122202)

摘　　要：现有目标跟踪算法一般是基于静止的相机平台展开的,为此提出一种相机运动状态下的实时运动目标跟踪算法。根据运动目标对光流结果的影响对特征点进行筛选,利用光流法评估相机的全局运动特征,参考评估全局运动的结果,修正粒子滤波的运动方程,其中选取颜色直方图设定为目标的特征模型,实现对移动目标的跟踪。实验结果表明,在相机运动的状态下,该算法能够快速精确跟踪运动目标,可以达到实时性要求,具有非常好的鲁棒性。Present object tracking algorithms are mainly implemented in static platforms. A novel technique for real-time object tracking in videos captured by cameras on mobile platforms was proposed. Firstly, feature points that have inconsistent optical flows with those of the background were ruled out. Secondly, optical flows on the rest feature points were used to estimate the global motion of the camera. Finally, the kinematic function of the particle filtering was modified by using the global motion of the camera, together with the color-space histogram as the appearance model, to achieve the robust tracking performance in unstable video sequences. The experimental results demonstrate that the proposed method achieves the promising real-time tracking performance and it can track the moving object robustly in videos captured by moving cameras.

关 键 词：移动平台 特征点筛选 全局运动估计 粒子滤波算法 目标跟踪 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]