Learning low-rank and discriminative dictionary for image classification

Dictionary learning plays a crucial role in sparse representation based image classification. In this paper, we propose a novel approach to learn a discriminative dictionary with low-rank regularization on the dictionary. Specifically, we apply Fisher discriminant function to the coding coefficients to make the dictionary more discerning, that is, a small ratio of the within-class scatter to between-class scatter. In practice, noisy information in the training samples will undermine the discriminative ability of the dictionary. Inspired by the recent advances in low-rank matrix recovery theory, we apply low-rank regularization on the dictionary to tackle this problem. The iterative projection method (IPM) and inexact augmented Lagrange multiplier (ALM) algorithm are adopted to solve our objective function. The proposed discriminative dictionary learning with low-rank regularization (D²L²R²) approach is evaluated on four face and digit image datasets in comparison with existing representative dictionary learning and classification algorithms. The experimental results demonstrate the superiority of our approach.     

Intra-class low-rank regularization for supervised and semi-supervised cross-modal retrieval

Cross-modal retrieval aims to retrieve related items across different modalities, for example, using an image query to retrieve related text. The existing deep methods ignore both the intra-modal and inter-modal intra-class low-rank structures when fusing various modalities, which decreases the retrieval performance. In this paper, two deep models (denoted as ILCMR and Semi-ILCMR) based on intra-class low-rank regularization are proposed for supervised and semi-supervised cross-modal retrieval, respectively. Specifically, ILCMR integrates the image network and text network into a unified framework to learn a common feature space by imposing three regularization terms to fuse the cross-modal data. First, to align them in the label space, we utilize semantic consistency regularization to convert the data representations to probability distributions over the classes. Second, we introduce an intra-modal low-rank regularization, which encourages the intra-class samples that originate from the same space to be more relevant in the common feature space. Third, an inter-modal low-rank regularization is applied to reduce the cross-modal discrepancy. To enable the low-rank regularization to be optimized using automatic gradients during network back-propagation, we propose the rank-r approximation and specify the explicit gradients for theoretical completeness. In addition to the three regularization terms that rely on label information incorporated by ILCMR, we propose Semi-ILCMR in the semi-supervised regime, which introduces a low-rank constraint before projecting the general representations into the common feature space. Extensive experiments on four public cross-modal datasets demonstrate the superiority of ILCMR and Semi-ILCMR over other state-of-the-art methods.

     

Discriminative low-rank preserving projection for dimensionality reduction

As an effective image clustering tool, low-rank representation (LRR) can capture the intrinsic representation of the observed samples. However, firstly, the good representation does not mean good classification performance. Secondly, no projection matrix is obtained in the training stage, and it cannot deal with the new samples. By incorporating the discriminant analysis and the local neighborhood relationship of the original samples into the low-rank representation, a novel discriminative low-rank preserving projection (DLRPP) algorithm is presented for dimensionality reduction. In DLRPP, the global structure information can be captured by LRR, and the local geometricinformation is simultaneously preserved by the manifold regularization term. The constrained term is induced by the adaptive graph, which is obtained by low-rank representation coefficients. In addition, by introducing discriminant analysis constraint term, DLRPP can learn an optimal projection matrix for data dimensionality reduction. The numerous experiments on six public image datasets prove that the proposed DLRPP can obtain better recognition accuracy compared with the state-of-the-art feature extraction methods.     

基于截断核范数低秩分解的自适应字典学习算法

针对过完备字典直接对图像进行稀疏表示不能很好地剔除高频噪声的影响,压缩感知后图像重构质量不高的问题,提出了基于截断核范数低秩分解的自适应字典学习算法。该算法首先利用截断核范数正则化低秩分解模型对图像矩阵低秩分解得到低秩部分和稀疏部分,其中低秩部分保留了图像的主要信息,稀疏部分主要包含高频噪声及部分物体轮廓信息;然后对图像低秩部分进行分块,依据图像块纹理复杂度对图像块进行分类;最后使用K奇异值分解（K-single value decomposition, K-SVD）字典学习算法,针对不同类别训练出多个不同大小的过完备字典。仿真结果表明,本文所提算法能够对图像进行较好的稀疏表示,并在很好地保持图像块特征一致性的同时显著提升图像重构质量。     

基于低秩稀疏评分的非监督特征选择

在处理高维数据过程中,特征选择是一个非常重要的数据降维步骤。低秩表示模型具有揭示数据全局结构信息的能力和一定的鉴别能力。稀疏表示模型能够利用较少的连接关系揭示数据的本质结构信息。在低秩表示模型的基础上引入稀疏约束项,构建一种低秩稀疏表示模型学习数据间的低秩稀疏相似度矩阵;基于该矩阵提出一种低秩稀疏评分机制用于非监督特征选择。在不同数据库上将选择后的特征进行聚类和分类实验,同传统特征选择算法进行比较。实验结果表明了低秩特征选择算法的有效性。     

A novel segmentation algorithm for nucleus in white blood cells based on low-rank representation

White blood cells (WBCs) segmentation is a challenging problem in the study of automated morphological systems, due to both the complex nature of the cells and the uncertainty that is present in video microscopy. This paper investigates how to boost the effects of region-based nucleus segmentation in WBCs by means of optimal thresholding and low-rank representation. The main idea is firstly using optimal thresholding to obtain the possible uniform WBC regions in the input image. After that, a manifold-based low-rank representation technique is employed to infer a unified affinity matrix that implicitly encodes the segmentation of the pixels of possible WBC regions. This is achieved by separating the low-rank affinities from the feature matrix into a pair of sparse and low-rank matrices. The experiments show that the proposed method is possible to produce better segmentation results compared with existing approaches.

     

Manifold Based Low-Rank Regularization for Image Restoration and Semi-Supervised Learning

Low-rank structures play important roles in recent advances of many problems in image science and data science. As a natural extension of low-rank structures for data with nonlinear structures, the concept of the low-dimensional manifold structure has been considered in many data processing problems. Inspired by this concept, we consider a manifold based low-rank regularization as a linear approximation of manifold dimension. This regularization is less restricted than the global low-rank regularization, and thus enjoy more flexibility to handle data with nonlinear structures. As applications, we demonstrate the proposed regularization to classical inverse problems in image sciences and data sciences including image inpainting, image super-resolution, X-ray computer tomography image reconstruction and semi-supervised learning. We conduct intensive numerical experiments in several image restoration problems and a semi-supervised learning problem of classifying handwritten digits using the MINST data. Our numerical tests demonstrate the effectiveness of the proposed methods and illustrate that the new regularization methods produce outstanding results by comparing with many existing methods.     

联合松弛块对角表示矩阵回归的遮挡人脸识别

基于核范数矩阵回归的方法能够有效解决人脸识别中连续遮挡的问题,然而该类方法仅关注误差图像的低秩结构信息,忽略了样本图像表示的相关性。为了有效解决自然场景下的遮挡人脸识别问题,考虑到这一特点,提出一种联合松弛块对角表示的矩阵回归模型（RBDMR）学习图像的松弛块对角表示,并通过动态优化表示矩阵的块对角分量加强类内表示的相关性和类间表示的差异性。此外,通过联合优化训练样本和测试样本的表示持续提高类内表示的一致性。通过在三个不同的数据集进行验证,实验结果表明,该方法优于其他对比算法,在真实遮挡和光照变化的情况下有较好的性能。     

基于非凸低秩分解判别的叠加线性稀疏人脸识别

目的 针对因采集的人脸图像样本受到污染而严重干扰人脸识别及训练样本较少（小样本）时会由于错误的稀疏系数导致性能急剧下降从而影响人脸识别的问题,提出了一种基于判别性非凸低秩矩阵分解的叠加线性稀疏表示算法。方法 首先由γ范数取代传统核范数,克服了传统低秩矩阵分解方法求解核范数时因矩阵奇异值倍数缩放导致的识别误差问题;然后引入结构不相干判别项,以增加不同类低秩字典间的非相干性,达到抑制类内变化和去除类间相关性的目的;最后利用叠加线性稀疏表示方法完成分类。结果 所提算法在AR人脸库中的识别率达到了98.67±0.57%,高于SRC（sparse representation-based classification）、ESRC（extended SRC）、RPCA（robust principal component analysis）+SRC、LRSI（low rank matrix decomposition with structural incoherence）、SLRC（superposed linear representation based classification）-l₁等算法;同时,遮挡实验表明,算法对遮挡图像具有更好的鲁棒性,在不同遮挡比例下,相比其他算法均有更高的识别率。在CMU PIE人脸库中,对无遮挡图像添加0、10%、20%、30%、40%的椒盐噪声,算法识别率分别达到90.1%、85.5%、77.8%、65.3%和46.1%,均高于其他算法。结论 不同人脸库、不同比例遮挡和噪声的实验结果表明,所提算法针对人脸遮挡、表情和光照等噪声因素依然保持较高的识别率,鲁棒性更好。     

基于鉴别性低秩表示的2阶段人脸识别算法

针对图像训练样本中存在噪声等情况,提出一种基于鉴别性低秩表示的2阶段人脸识别算法。该算法第1阶段是对所有训练样本进行低秩处理,筛选出M类与测试样本最相近的样本用于粗分类;第2阶段使用第1阶段筛选出来的样本做鉴别性低秩表示处理,并使用稀疏线性表示进行精细分类,决定测试样本最适合的类标签。本算法结合了低秩算法与稀疏算法的优点,在标准人脸库上的实验表明本算法表现优越。     

基于低秩矩阵二元分解的快速显著性目标检测算法

近年来,基于矩阵低秩表示模型的图像显著性目标检测受到了广泛关注。在传统模型中通常对秩最小化问题进行凸松弛,即引入最小化核范数将原始输入图像分解为低秩矩阵和稀疏矩阵。但是,这种方法在每次迭代中必须执行矩阵奇异值分解（SVD）,计算复杂度较高。为此,本文提出了一种低秩矩阵双因子分解和结构化稀疏矩阵分解联合优化模型,并应用于显著性目标检测。算法不仅利用低秩矩阵双因子分解和交替方向法（ADM）来降低时间开销,而且引入分层稀疏正则化刻画稀疏矩阵中元素之间的空间关系。此外,所提算法能够无缝集成高层先验知识指导矩阵分解过程。实验结果表明,提出模型和算法的检测性能优于当前主流无监督显著性目标检测算法,且具有较低的时间复杂度。     

结构化低秩字典学习的人脸识别

目的 现实中采集到的人脸图像通常受到光照、遮挡等环境因素的影响,使得同一类的人脸图像具有不同程度的差异性,不同类的人脸图像又具有不同程度的相似性,这极大地影响了人脸识别的准确性。为了解决上述问题对人脸识别造成的影响,在低秩矩阵恢复理论的基础上提出了具有识别力的结构化低秩字典学习的人脸识别算法。方法 该算法基于训练样本的标签信息将低秩正则化以及结构化稀疏同时引入到学习的具有识别力的字典上。在字典学习过程中,首先利用样本的重建误差约束样本与字典之间的关系;其次将Fisher准则应用到稀疏编码过程中,使其编码系数具有识别能力;由于训练样本中的噪声信息会影响字典的识别力,所以在低秩矩阵恢复理论的基础上将低秩正则化应用到字典学习过程中;接着,在字典学习过程中加入了结构化稀疏使其不丢失结构信息以保证对样本进行最优分类;最后再利用误差重构法对测试样本进行分类识别。结果 本文算法在AR以及ORL人脸数据库上分别进行了实验仿真。在AR人脸数据库中,为了分析样本不同维数对实验结果造成的影响,选取了第一时期拍摄的每人6幅图像,包括1幅围巾遮挡,2幅墨镜遮挡以及3幅脸部表情变化以及光照变化（未被遮挡）的图像作为训练样本,同时选取相同组合的样本图像作为测试样本,无论哪种方法,图像的维度越高识别率越高。对比SRC （sparse representation based on classification）算法与DKSVD （discriminative K-means singular value decomposition）算法的识别率可知,DKSVD算法通过字典学习减缓了训练样本中的不确定因素对识别结果的影响;对比DLRD_SR （discriminative low-rank dictionary learning for sparse representation）算法与FDDL （Fisher discriminative dictionary learning）算法的识别率可知,当图像有遮挡等噪声信息存在时,字典低秩化可以提高至少5.8%的识别率;对比本文算法与DLRD_SR算法可知,在字典学习的过程中加入Fisher准则后识别率显著提高,同时理想稀疏值能保证对样本进行最优的分类。当样本图像的维度达到500维时人脸图像在有围巾、墨镜遮挡的情况下识别率可达到85.2%;其中墨镜和围巾的遮挡程度分别可以看成是人脸图像的20%和40%,为了验证本文算法在不同脸部表情变化、光照改变以及遮挡情况下的有效性,根据训练样本的具体图像组合情况进行实验。无论哪种样本图像组合,本文算法在有遮挡存在的样本识别中具有显著优势。在训练样本只包含脸部表情变化、光照变化以及墨镜遮挡图像的情况下,本文算法的识别率高于其他算法至少2.7%,在训练样本只包含脸部表情变化、光照变化以及围巾遮挡图像的情况下,本文算法的识别率高于其他算法至少3.6%,在训练样本包含脸部表情变化、光照变化、围巾遮挡以及墨镜遮挡图像的情况下,其识别率高于其他算法至少1.9%。在ORL人脸数据库中,人脸图像在无遮挡的情况下识别率达到95.2%,稍低于FDDL算法的识别率;在随机块遮挡程度达到20%时,相比较于SRC算法、DKSVD算法、FDDL算法以及DLRD_SR算法,本文算法的识别率最高;当随机块遮挡程度达到50%时,以上算法的识别率均不高,但本文算法的其识别率仍然最高。结论 本文算法在人脸图像受到遮挡等因素的影响时具有一定的鲁棒性,实验结果表明该算法在人脸识别方面具有可行性。     

基于结构化低秩恢复的鲁棒人脸识别算法

由于数据本身的自表示特性，当给定一个字典时，同类样本理论上具有相似的线性表示，所以所有样本的表示矩阵具有块对角结构。但在由于样本中存在的各种污损，数据子空间结构可能会被破坏。为了解决这一问题，很多基于低秩表示的恢复算法相继提出，但是仅有对表示的低秩约束并不能很好地将原始训练样本转化到理想的低秩子空间。因此，提出了一个鲁棒的结构化低秩恢复算法（Robust Structured Low-Rank Recovery，RSLRR）。RSLRR利用理想的标签矩约束阵促进低秩表示趋近于块对角结构，以此挖掘更多的潜在结构信息。同时，为了减少严格的趋近0-1标签矩阵造成的结构信息损失，RSLRR增加了一个正则化项用来减弱非块对角系数的负面影响。通过RSLRR算法可以得到一个判别的结构化字典，并可计算出一个低秩投影矩阵将所有测试样本有效的投影到其相应的低秩子空间。在AR和CMU PIE数据库上的实验结果验证了RSLRR算法的有效性和鲁棒性。     

基于半监督典型相关分析的多视图维数约简

为了有效地在半监督多视图情景下进行维数约简,提出了使用非负低秩图进行标签传播的半监督典型相关分析方法。非负低秩图捕获的全局线性近邻可以利用直接邻居和间接可达邻居的信息维持全局簇结构,同时,低秩的性质可以保持图的压缩表示。当无标签样本通过标签传播算法获得估计的标签信息后,在每个视图上构建软标签矩阵和概率类内散度矩阵。然后,通过最大化不同视图同类样本间相关性的同时最小化每个视图低维特征空间类内变化来提升特征鉴别能力。实验表明所提方法比已有相关方法能够取得更好的识别性能且更鲁棒。     

针对混合污染的结构化鲁棒低秩恢复算法在人脸识别中的应用

传统的低秩恢复算法在识别有混合污染的人脸图像时,通常只对污染部分进行一种类型的约束,并不能很好地恢复出干净的样本。针对这种情况,提出结构化鲁棒低秩恢复算法（structured and robust low-rank recovery for mixed contamination,SRLRR）。SRLRR算法利用对二维误差图像的低秩约束移除样本中的连续污染部分,同时利用稀疏约束分离样本中服从拉普拉斯分布的噪声。另外,为了学习到更具有鉴别性的低秩表示,该算法对表示系数进行了块对角结构化约束。在三个常用数据库上的实验证明了SRLRR算法的有效性和鲁棒性。     

Low Rank Prior and Total Variation Regularization for Image Deblurring

The similar image patches should have similar underlying structures. Thus the matrix constructed from stacking the similar patches together has low rank. Based on this fact, the nuclear norm minimization, which is the convex relaxation of low rank minimization, leads to good denoising results. Recently, the weighted nuclear norm minimization has been applied to image denoising. This approach presents state-of-the-art result for image denoising. In this paper, we further study the weighted nuclear norm minimization problem for general image recovery task. For the weights being in arbitrary order, we prove that such minimization problem has a unique global optimal solution in the closed form. Incorporating this idea with the celebrated total variation regularization, we then investigate the image deblurring problem. Numerical experimental results illustratively clearly that the proposed algorithms achieve competitive performance.     

基于lP范数的非凸低秩张量最小化

在低秩矩阵、张量最小化问题中,凸函数容易求得最优解,而非凸函数可以得到更低秩的局部解.文中基于非凸替换函数的低秩张量恢复问题,提出基于l_p范数的非凸张量模型.采用迭代加权核范数算法求解模型,实现低秩张量最小化.在合成数据和真实图像上的大量实验验证文中方法的恢复性能.     

基于低秩表示与矩阵填充的人脸识别方法

基于回归分析的人脸识别方法在处理不完备数据矩阵时,先对矩阵进行填充,再使用人脸识别方法,因此会降低分类性能.为了更有效地执行关于不完备数据的识别,文中将低秩矩阵填充和低秩表示学习整合在同一个模型,提出基于低秩表示和低秩矩阵填充的人脸识别方法.通过最小化表示系数和矩阵秩交替计算样本低秩表示系数矩阵和恢复矩阵缺失项,再使用最近邻分类器实现分类.在一些公开人脸数据集上的实验表明,在训练样本矩阵元素随机缺失时,文中方法可以有效提高识别精度及降低填充误差.     

结构图正则低秩子空间聚类

针对结构稀疏子空间聚类中不能很好地保证相似度矩阵连接性的问题,给出了一个新的统一优化模型。首先,引入了表示系数矩阵的子空间结构范数,增加了低秩表示来揭示高维数据的全局结构。其次,为了使相似度矩阵具有类内统一,类间稀疏的作用,还定义了分组效应来捕获数据的内部几何结构,提出了结构图正则低秩子空间聚类模型。最后使用自适应惩罚的线性化交替法（LADMAP）来得到最优解。实验结果表明,该模型不但可以捕获数据的全局结构,而且还可以捕获数据的内在几何结构,迫使相关数据紧密结合,不相关数据松散分离,从而使得相似度矩阵与分割矩阵变得更加一致。     

改进稀疏表示算法在人脸识别中的应用

人脸识别的主要难度在于，受到光照变化、表情变化以及遮挡的影响，会使得采集的不同人的人脸图像具有相似性。为有效解决基于稀疏表示的分类算法（Sparse Representation-based Classification，SRC）在人脸训练样本不足时会导致识别率降低和稀疏表示求解效率较低的问题，提出了基于判别性低秩分解与快速稀疏表示分类（Low Rank Recovery Fast Sparse Representation-based Classification，LRR_FSRC）的人脸识别算法。利用低秩分解理论得到低秩恢复字典以及稀疏误差字典，结合低秩分解和结构不相干理论，训练出判别性低秩类字典和稀疏误差字典，并把它们结合作为测试时所用的字典；用坐标下降法来求解稀疏系数以提高了计算效率；根据重构误差实现测试样本的分类。在YALE和ORL数据库上的实验结果表明，提出的基于LRR_FSRC的人脸识别方法具有较高的识别率和计算效率。