摘要
为克服二维主成分分析(2DPCA)跟踪效率低的缺点,提出一种基于双向二维主成分分析(Bi-2DPCA)的运动目标跟踪算法。采用双向二维主成分分析作为目标表示的方法建立目标图像子空间,同时在图像均值与协方差矩阵的更新中引入基于目标图像匹配程度的自适应增量因子的增量学习的方法进一步提高算法效率。在多个包含动态背景的图像序列上的对比实验结果表明算法能在目标处于部分遮挡的情况下准确跟踪目标,同时算法在效率上高于基于二维主成分分析的目标跟踪算法。
An object tracking algorithm based on bidirectional two-dimensional principle component analysis （Bi-2DPCA） is proposed. Object representation based on Bi-2DPCA is used to generate the object image subspace. To increase the speed of al- gorithm, an incremental learning method based on proposed adaptive incremental factor according to the object image match de- gree is adopted to update the related mean matrix and covariance matrices. The comparative experiments on classical image se- quences containing dynamic backgrounds have been carried out, the results show the proposed algorithm is capable of tracking object accurately even in ease of partial occlusion, and more efficient than the algorithm based on two-dimensional principle component analysis.
出处
《计算机工程与应用》
CSCD
2013年第22期155-159,共5页
Computer Engineering and Applications
基金
安徽省科技攻关强警专项(No.1101b0403030)
国家自然科学基金(No.61271352)
中国科学院上海微系统与信息技术横向研发基金课题资助
关键词
二维主成分分析
双向二维主成分分析
目标跟踪
增量学习
two-dimensional principle component analysis
bidirectional two-dimensional principle component analysis
objecttracking
incremental leaming
