软判决稀疏字典的目标跟踪算法  

Object Tracking with Soft Discriminative Sparse Dictionary

作　　者：查绎 曹铁勇[1] 黄辉[1] 宋智军[2] 尤峻[1] 

机构地区：[1]解放军理工大学指挥信息系统学院,南京210007 [2]中国电子科技集团公司第二十八所,南京210007

出　　处：《计算机辅助设计与图形学学报》2014年第8期1279-1289,共11页Journal of Computer-Aided Design & Computer Graphics

摘　　要：为提高视频目标跟踪算法的鲁棒性,提出一种基于在线更新稀疏模板的自适应参数特征判别跟踪算法.该算法采用离线方式训练出基于方向梯度直方图特征的字典,用于目标表示和线性分类器训练,从而构建出非固定参数的观测模型;观测模型中动态调整的权重系数由采用正负模板构建形成的稀疏字典进行实时动态更新;将观测模型与粒子滤波相结合对当前帧的各候选采样进行观测,得出跟踪结果.实验结果表明,文中算法具有相对较好的鲁棒性.In this paper,a soft discriminative tracking method with adaptive parameters is proposed which is based on online sparse template.Firstly,an off-line local-constrained dictionary based on HOG is used to represent the object,which can make this tracking algorithm more robust.Then a dynamic parameter observation model is constructed with online sparse dictionary and the linear classifier which is learned from the local-constrained dictionary.In the end,particle filter is used to estimate the states of the object.The experimental results on several challenging image sequences demonstrate that the proposed tracker achieves more favorable performance than the state-of-art methods.

关 键 词：软判决 稀疏模板 目标跟踪 鲁棒性 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]