基于区域协方差矩阵和2DPCA学习的视频跟踪方法研究  

Visual Object Tracking Algorithm Based on Region Covariance Matrix and 2DPCA Learning

作　　者：张焕龙[1,2] 郑卫东[2] 舒云星[2] 蒋斌[3] 

机构地区：[1]上海交通大学航空航天学院,上海200240 [2]洛阳理工学院计算机与信息工程系,洛阳471003 [3]郑州轻工业学院计算机与通信工程学院,郑州450002

出　　处：《计算机科学》2015年第9期278-281,共4页Computer Science

基　　金：国家自然科学基金项目(61503173)资助

摘　　要：针对PCA在视频跟踪应用中需要将图像转换成向量而造成信息丢失和小样本等问题,提出一种基于2DPCA学习的自适应性视频跟踪方法。该方法将图像矩阵直接进行处理,保持了跟踪目标的空间结构信息。在粒子滤波框架下采用仿射变换运动模型,并通过协方差特征融合方式评估目标运动状态,提高了目标外观模型的学习能力,实现了鲁棒的自适应性跟踪效果。进行了标准的视频序列测试,结果证明提出的算法能够较好地适应目标姿态、光线和部分遮挡等跟踪问题。Against the information loss problem during the process of transferring image into vector using PCA in visual tracking, a new adaptive object tracking method based on 2DPCA learning was proposed. It takes the tracked obiect as a matrix, which can maintain the spatial structure information of the target. And in the particle filter framework, the algo- rithm adopts the affine model to describe object motion. Meanwhile, in order to enhance the learning ability, the algo- rithm uses covariance feature fusion to estimate the motion states of the tracked object so as to obtain the robust track- ing results. Experimental results indicate that the proposed method achieves favorable performance when the object un- dergoes illumination changes, pose changes,and partial occlusion between consecutive frames.

关 键 词：2DPCA跟踪 仿射变换 协方差特征融合 特征基学习 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]