Low-Rank Embedding for Robust Image Feature Extraction
Low-Rank Embedding for Robust Image Feature Extraction
复制标题
DOI：
10.1109/tip.2017.2691543
复制
发表时间：
2017-06
影响因子：
10.6
通讯作者：
W. Wong;Zhihui Lai;J. Wen;Xiaozhao Fang;Yuwu Lu
中科院分区：
文献类型：
--
作者：
W. Wong;Zhihui Lai;J. Wen;Xiaozhao Fang;Yuwu Lu
对噪声、离群点和污染的稳健性是线性降维中的一个重要问题。由于样本特有的腐败和离群点的存在，破坏了类特有的结构或局部几何结构，因此，现有的许多方法，包括流行的基于流形学习的线性维度方法，在识别任务中都不能取得良好的性能。在本文中，我们通过引入稳健的低阶表示(LRR)来研究无监督的稳健线性降维。为此，本文提出了一种称为低阶嵌入(LRE)的稳健线性降维技术，它提供了一种稳健的图像表示来揭示图像之间的潜在关系，以减少遮挡和损坏的负面影响，从而增强了算法在图像特征提取中的稳健性。LRE同时搜索最优LRR和最优子空间。LRE模型可以通过交替迭代变元拉格朗日乘子法和特征分解法来求解。给出了算法的理论分析，包括收敛分析和计算复杂度。在一些具有不同损坏的知名数据库上的实验表明，LRE优于以往的特征提取方法，从而表明了该方法的健壮性。本文的代码可以从http://www.scholat.com/laizhihui.下载
Robustness to noises, outliers, and corruptions is an important issue in linear dimensionality reduction. Since the sample-specific corruptions and outliers exist, the class-special structure or the local geometric structure is destroyed, and thus, many existing methods, including the popular manifold learning- based linear dimensionality methods, fail to achieve good performance in recognition tasks. In this paper, we focus on the unsupervised robust linear dimensionality reduction on corrupted data by introducing the robust low-rank representation (LRR). Thus, a robust linear dimensionality reduction technique termed low-rank embedding (LRE) is proposed in this paper, which provides a robust image representation to uncover the potential relationship among the images to reduce the negative influence from the occlusion and corruption so as to enhance the algorithm’s robustness in image feature extraction. LRE searches the optimal LRR and optimal subspace simultaneously. The model of LRE can be solved by alternatively iterating the argument Lagrangian multiplier method and the eigendecomposition. The theoretical analysis, including convergence analysis and computational complexity, of the algorithms is presented. Experiments on some well-known databases with different corruptions show that LRE is superior to the previous methods of feature extraction, and therefore, it indicates the robustness of the proposed method. The code of this paper can be downloaded from http://www.scholat.com/laizhihui.