在线低秩稀疏表的鲁棒视觉跟踪.docVIP

MACROBUTTON AcceptAllChangesInDoc 在线低秩稀疏表示的鲁棒视觉跟踪 孔繁锵，王丹丹，沈 秋，卞陈鼎，严小乐 （南京航空航天大学 航天学院，江苏 南京 210016） 摘 要：基于L1最小化的鲁棒视觉跟踪算法(L1跟踪算法)使用图像灰度值特征描述目标，忽略了模板间的结构信息，对目标外观变化的建模不够准确，导致跟踪准确度较低。而且L1跟踪算法为了平衡跟踪速度和跟踪效果而采用分辨率较低的1215图像块，难以获取足够的信息来表征目标。针对L1跟踪算法的不足，该文提出一种在线低秩稀疏表示的视觉跟踪算法。首先，该算法充分利用主成分分析(PCA)基向量对目标外观变化的表示能力并考虑目标遮挡现象，以PCA基向量模板描述目标外观变化，以琐碎模板处理遮挡等异常噪声，从而将候选目标表示为PCA基模板和琐碎模板的线性组合。其次在目标表示模型的优化问题中，对PCA基模板系数进行低秩约束和L1,1范数正则化约束，对琐碎模板系数实施L1,1范数约束，并采用非精确增广拉格朗日乘子方法(IALM)求解表示系数。然后在粒子滤波框架下，用目标未被遮挡部分的重建误差和稀疏误差项建立观测模型跟踪目标。最后为了克服模型漂移问题，采用遮挡检测更新机制进行模板更新。在对8组视频图像序列进行测试的实验中，图像块分辨率设定为3232，与4个现有的跟踪算法相比，该算法取得了最高的平均重叠率0.78和最低的平均中心误差4.05。实验结果表明，该文提出的跟踪算法具有较好的跟踪准确性和鲁棒性。 关键词：视觉跟踪；低秩表示；稀疏表示；PCA基向量 中图分类号：TP391.41 文献标识码：A Robust Visual Tracking via Online Low-rank Sparse Representation KONG Fanqiang, WANG Dandan, SHEN Qiu, BIAN Chending，YAN Xiaole (College of Astronautics, Nanjing Univ. of Aeronautics and Astronautics, Nanjing 210016, China) Abstract: Image gray value feature was used to describe the target in robust visual tracking via L1 minimization (L1 tracking algorithm), ignoring the structural information between the templates, modeling inaccurate changes of target appearance, leading to a low tracking accuracy. What’s more, since the low resolution image block with 1215 was adopted to balance the tracking speed and tracking effect, it was difficult to obtain sufficient information to represent the target. To solve the problem of the L1 tracking algorithm, a visual tracking method based on online low-rank sparse representation was proposed. Firstly, in order to take advantage of the ability to represent target appearance changes of Principal Component Analysis (PCA) basis vector and take occlusion into account, the PCA basis vector template was used to describe target appearance changes, and the occlusion and other abnormal noise were modeled with the trivial template. Therefore, the candidate target was represented as a linear combination of PCA basis templates and triv