基于鉴别性低秩表示及字典学习的鲁棒人脸识别算法

针对人脸识别中的图像存在噪声等情况,提出基于鉴别性低秩表示及字典学习的算法。使用鉴别性低秩子空间恢复算法(discriminative low-rank representation, DLRR)获得类别间尽可能独立且干净的训练样本,然后通过引入基于Fisher准则的字典学习(Fisher Discrimination Dictionary Learning, FDDL)方法得到结构化字典,其子字典对对应的类有较好的表示能力,约束编码系数具有较小类内散列度和较大类间散列度。最后对测试样本稀疏线性表示时正确类别的样本贡献更大。在标准人脸数据库上的实验结果表明该算法有较好性能。     

基于低秩分解的鲁棒典型相关分析

典型相关分析（CCA）是一种经典的多特征提取算法,它能够有效地抽取两组特征之间的相关性,现已被广泛应用于模式识别。在含噪声数据情况下,CCA的特征表示性能受到限制。为了使CCA更好地处理含噪声数据,提出一种基于低秩分解的典型相关分析算法——鲁棒典型相关分析（robust canonical correlation analysis,RbCCA）。RbCCA首先对特征集进行低秩分解,得到低秩分量和噪声分量,以此分别构建对应的协方差矩阵。通过最大化低秩分量的相关性,同时最小化噪声分量的相关性来建立判别准则函数,进而求取鉴别投影矢量。在MFEAT手写体数据库、ORL和Yale人脸数据中的实验结果表明,在包含噪声的情况下,RbCCA的识别效果优于现有的典型相关分析方法。     

多矩阵低秩分解的鲁棒特征提取

由于传统的人脸识别算法效果容易受制于光照、表情、遮挡以及稀疏大噪声等外界因素的影响,如何有效提取数据特征、进一步提升算法的鲁棒性,是传统人脸识别方法发展的关键所在.本文将多矩阵低秩分解应用在人脸特征提取中,充分利用多张人脸之间的结构相似性,探索人脸图像集的低秩子空间,进而结合低秩矩阵恢复模型来提取测试样本的低秩特征.最...     

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.     

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。     

基于结构化低秩表示和低秩投影的人脸识别算法

在实际的人脸识别中,给定的训练图像往往存在遮挡和噪声,导致稀疏表示分类（SRC）算法的性能下降。针对上述问题,提出一种基于结构化低秩表示（SLR）和低秩投影的人脸识别方法--SLR_LRP。首先通过SLR对原始训练样本进行低秩分解得到干净的训练样本,根据原始训练样本和恢复得到的干净训练样本得到一个低秩投影矩阵;然后将测试样本投影到该低秩投影矩阵;最后使用SRC对恢复后的测试样本进行分类。在AR人脸库和Extended Yale B人脸库上的实验结果表明,SLR_LRP可以有效处理样本中存在的遮挡和像素破坏。     

基于特征化字典的低秩表示人脸识别

针对现有低秩表示（LRR）算法中全局与局部人脸特征信息融合不足的问题,提出了一种新的人脸识别算法——基于特征化字典的低秩表示（LRR-CD）。首先,将每张人脸照片表示成一个个特征化字典的集合,然后同时最小化基于训练样本的低秩重构特征系数以及与之相对应的类内特征差异。为了获得高效且具有高判别性的人脸图像的特征块重构系数矩阵,提出了一种新的数学公式模型,通过同时求解训练样本中相对应的特征块以及对应的类内特征差异词典的低秩约束问题,尽可能完整地保留原始高维人脸图像中的全局和局部信息,尤其是局部类内差异特征。另外,由于对特征块中信息的充分挖掘,所提算法对于一般程度上的面部遮挡和光照等噪声影响具有良好的鲁棒性。在AR、CMU-PIE和Extended Yale B人脸数据库进行多项对比实验,由实验结果可知LRR-CD相较于对比的稀疏表示（SRC）、协从表示（CRC）、低秩表示正规切（LRR-NCUT）和低秩递归最小二乘（LRR-RLS）算法在平均识别率上有2.58~17.24个百分点的提高。实验结果表明LRR-CD性能优于与之对比的算法,可以更高效地用于人脸全局和局部特征信息的融合,且具有优良的识别率。     

基于低秩表示线性保持投影的特征提取算法

为了在特征提取过程中保持数据低秩特性不变,提出了一种基于低秩表示的线性保持投影算法用于维数约简。它能够使降维后的低维空间中的数据依旧较好地保持在原始高维空间中的低秩特性,准确地学习出数据的低维子空间。通过构建两个不同的低秩表示模型来 揭示两种不同结构特性的低秩权重,然后以保持数据的这两个低秩权重关系为目的来求解高维数据的低维空间。 在ORL库和Yale库人脸库上的实验结果证明,该算法比传统的特征提取方法更有效。     

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。     

基于鲁棒低秩张量恢复的高光谱图像去噪

去噪是高光谱图像进一步分析的重要预处理步骤,许多去噪方法都被用于高光谱图像数据立方体的去噪.然而,传统的去噪方法对异常值和非高斯噪声很敏感.文中利用底层干净H SI的张量性质数据、异常值的稀疏性质和非高斯噪声,提出一个新的基于鲁棒低秩张量修复的模型,从而在保护H SI的同时删除离散值的全局结构和不同类型的噪声(高斯噪声、脉冲噪声、死线等).该模型可以用非精确增广拉格朗日法求解,仿真和真实高光谱图像实验的结果表明,该方法对H SI去噪是有效的.     

一种新的低秩分块稀疏表示的人脸识别算法

针对 人脸图像中表情变化、遮挡、光照的问题,本文提出了一种新颖的基于低秩分块稀疏表示的 人脸识别算法。该算法采用了一种新的结构不相关的低秩矩阵恢复方法,同时采用离散余弦 变换方法联合处理人脸图像中遮挡、掩饰和光照的问题,对处理过的图片采用一种独特的重 叠分块方法,利用冗余信息有效地提高了算法的识别率。在分类阶段,利用Alignment pool ing的方法,有效地提高了识别速度。该算法在标准人脸数据库上进行了多次实验,实验结 果表明：与现有人脸识别算法相比,算法的识别准确率和计算效率都得到了一致提高。     

Face Recognition Using A Low Rank Representation Based Projections Method

In this paper, a low rank representation based projections (LRRP) method is presented for face recognition. In LRRP, low rank representation is used to construct a nuclear graph to characterize the local compactness information by designing the local scatter matrix like SPP; the total separability information is characterized by the total scatter like PCA. LRRP seeks the projection matrix simultaneously maximizing the total separability and the local compactness. Experimental results on FERET, AR, Yale face databases and the PolyU finger-knuckle-print database demonstrate that LRRP works well for face recognition.     

SAR Sea Ice Image Segmentation Method based on Low Rank Sparse Representation and Improved MRF Model

Accurate segmentation of Synthetic Aperture Radar (SAR)images is the premise of interpreting the distribution information of sea ice.However the existing segmentation methodsare seriously interfered by speckle noise,which leads to high segmentation error and low reliability interpreting results.In this paper,a novel sea ice SAR image segmentation method based on low rank sparse representation is proposed,firstly sparse components are extracted from the source image by using robust principal component,and then bilateral filter is used to enhance the image details.Due to the MRF segmentation model based on fixed potential function cannot accurately reflect the relevance between the areas,MRF segmentation model based on interactive potential function is built to segment the sea ice image accurately.A series of Radarsat satellites data are tested to validate performance of the proposed method,the results show that compare with traditional segmentation algorithms,the proposed method algorithm can not only maintain the connectivity of the image better,but also has higher segmentation accuracy.     

基于低秩矩阵补全的视频去噪研究

针对视频数据中严重的混合噪声问题，提出了一种基于块的视频去噪算法。通过对空间域和时间域的相似块进行分组，将混合噪声的去噪问题转化为一个低秩矩阵补全问题，从而得到一个对噪声统计特征没有强假设的去噪方法。由此产生的核范数最小化问题通过拉格朗日函数和不动点迭代算法得到有效的解决。实验验证了所提出的视频去噪方法在去除混合噪声方面的鲁棒性和有效性。     

Robust Optical Flow Computation Based on Least-Median-of-Squares Regression

An optical flow estimation technique is presented which is based on the least-median-of-squares (LMedS) robust regression algorithm enabling more accurate flow estimates to be computed in the vicinity of motion discontinuities. The flow is computed in a blockwise fashion using an affine model. Through the use of overlapping blocks coupled with a block shifting strategy, redundancy is introduced into the computation of the flow. This eliminates blocking effects common in most other techniques based on blockwise processing and also allows flow to be accurately computed in regions containing three distinct motions.A multiresolution version of the technique is also presented, again based on LMedS regression, which enables image sequences containing large motions to be effectively handled.An extensive set of quantitative comparisons with a wide range of previously published methods are carried out using synthetic, realistic (computer generated images of natural scenes with known flow) and natural images. Both angular and absolute flow errors are calculated for those sequences with known optical flow. Displaced frame difference error, used extensively in video compression, is used for those natural scenes with unknown flow. In all of the sequences tested, a comparison with those methods that result in a dense flow field (greater than 80% spatial coverage), show that the LMedS technique produces the least error irrespective of the error measure used.     

A Robust Audio Watermark Representation Based on Linear Chirps

In this paper, we introduce a novel watermark representation for audio watermarking, where we embed linear chirps as watermark signals. Different chirp rates, i.e., slopes on the time–frequency (TF) plane, represent watermark messages such that each slope corresponds to a unique message. These watermark signals, i.e., linear chirps, are embedded and extracted using an existing watermarking algorithm. The extracted chirps are then postprocessed at the receiver using a line detection algorithm based on the Hough–Radon transform (HRT). The HRT is an optimal line-detection algorithm, which detects directional components that satisfy a parametric constraint equation in the image of a TF plane, even at discontinuities corresponding to bit errors. Simulation results show that HRT correctly detects the embedded watermark message after common signal processing operations for bit error rates up to 20%. The new watermark representation and the postprocessing stage based on HRT significantly improve the performance of the watermark detection process and can be combined with existing watermark embedding/extraction algorithms for increased robustness.     

基于低秩模型的电力能源大数据异常修正方法研究

传统的电力能源大数据异常修正方法存在搜索次数过多问题,会造成异常数据辨识结果异常、修正结果不准确.为此,引入低秩模型,改善以上问题.采用低秩模型处理电力能源数据样本,去除样本数据噪声;在离线模式下,通过训练支持向量机对数据样本进行聚类;在联机模式下,利用滑动窗口辨识异常数据;针对单个或多个不相关的异常数据,依据基尔霍夫...     

基于稳健回归的AFM图像水平矫正算法

 AFM（Atomic Force Microscope,原子力显微镜）图像经常会出现背景倾斜或弯曲。背景倾斜的原因源于探针和样本表面的倾角或XYZ扫描仪带来的弯曲。本文将稳健的MM估计算法应用到AFM图像二维背景拟合中,消除背景的倾斜,并利用fast-s估计算法作为初始化,以缩短计算时间。实验结果表明,与传统方法相比,本方法的AFM图像水平矫正效果更好。     

Robust Spectral Subspace Clustering Based on Least Square Regression

In recent years, graph based subspace clustering has attracted considerable attentions in computer vision, as its capability of clustering data efficiently. However, the graph weights built by using representation coefficients are not the exact ones as the traditional definition. That is, the two steps are conducted in independent manner such that an overall optimal result cannot be guaranteed. To this end, in this paper, a novel subspace clustering via learning an adaptive graph affinity matrix is proposed, where the soft label and the representation coefficients of data are learned in an unified framework. First, the proposed method learns a robust representation for the data through least square regression, which reveals the subspace structure within data and captures various noises inside. Second, the segmentation is sought by conducting spectral clustering simultaneously. Most importantly, during the optimization process, the segmentation is utilized to iteratively enhance the block-diagonal structure of the learned representation to further assist the clustering process. Experimental results on several famous databases demonstrate that the proposed method performs better against the state-of-the-art approaches, in clustering.