Face recognition using recursive Fisher linear discriminant.

Fisher linear discriminant (FLD) has recently emerged as a more efficient approach for extracting features for many pattern classification problems as compared to traditional principal component analysis. However, the constraint on the total number of features available from FLD has seriously limited its application to a large class of problems. In order to overcome this disadvantage, a recursive procedure of calculating the discriminant features is suggested in this paper. The new algorithm incorporates the same fundamental idea behind FLD of seeking the projection that best separates the data corresponding to different classes, while in contrast to FLD the number of features that may be derived is independent of the number of the classes to be recognized. Extensive experiments of comparing the new algorithm with the traditional approaches have been carried out on face recognition problem with the Yale database, in which the resulting improvement of the performances by the new feature extraction scheme is significant.     

Subclass linear discriminant analysis for video-based face recognition

We present a novel subclass Linear Discriminant Analysis algorithm for feature extraction that copes with the severe pose, expression and illumination changes present in faces extracted from far-field video streams with subjects unconstrained in their motion and uncooperative to the system. Our novelty lies on the efficient automatic generation of subclasses from the gallery faces, by exploiting their different visual appearance and not constrained by their numbers per class. The proposed feature extraction algorithm is integrated in our complete face recognition system, with modules for preprocessing, classification, and decision fusion. We demonstrate the capability of the new algorithm to automatically generate discriminable subclasses and the resulting improved classification accuracy on a challenging video-based dataset, comprising low quality and resolution faces, as well as large variations in visual appearance. Our results indicate superior recognition rate compared to any systems in the CLEAR 2007 evaluation, running on that dataset.     

Keystroke features recognition based on stable linear discriminant analysis

A novel keystroke features recognition method based on stable linear discriminant Analysis (SLDA) was put forward.First of all,it maximum the dispersion between different sequences,while minimizing the dispersion between the same sequence set,maintain the best discriminant characteristics of the keystroke sequences.Secondly,the local similarity graph between keystroke sequences is constructed,minimizing the dispersion of the local similarity sequences,to keep the local similarity of keystroke sequences.Finally,based on the principles above,the feature of keystroke sequences are extracted,and the nearest neighbor classification criterion is used to judge the outputs.The effectiveness of the proposed method is certified by experiment results.     

Feature extraction using recursive cluster-based linear discriminant with application to face recognition.

A novel recursive procedure for extracting discriminant features, termed recursive cluster-based linear discriminant (RCLD), is proposed in this paper. Compared to the traditional Fisher linear discriminant (FLD) and its variations, RCLD has a number of advantages. First of all, it relaxes the constraint on the total number of features that can be extracted. Second, it fully exploits all information available for discrimination. In addition, RCLD is able to cope with multimodal distributions, which overcomes an inherent problem of conventional FLDs, which assumes uni-modal class distributions. Extensive experiments have been carried out on various types of face recognition problems for Yale, Olivetti Research Laboratory, and JAFFE databases to evaluate and compare the performance of the proposed algorithm with other feature extraction methods. The resulting improvement of performances by the new feature extraction scheme is significant.     

基于子模式双向二维线性判别分析的人脸识别

针对表情和光照变化等对人脸识别影响的问题,提出一种基于子模式双向二维线性判别分析(Sub-pattern two-directional two-dimensional linear discriminant analysis,Sp-(2D)2 LDA)的人脸识别方法。该方法首先对原图像进行分块处理,并保持子块间的空间关系,然后对各个子训练样本集从行方向和列方向同时利用2DLDA进行特征抽取,最后把各个子特征矩阵拼接成一对应原始图像的特征矩阵,并采用最近邻分类器进行分类识别。在ORL及Yale人脸库上的试验结果表明,Sp-(2D)2 LDA有效降低了鉴别特征的维数,减少了表情和光照变化的影响,获得了较好的识别性能。     

基于块双向Fisher线性判别分析人脸识别

为解决二维Fisher线性判别(2DFLD)分析需要较多系数用以表示图像的特征阵、只考虑了图像的列间相关性从而忽略行间相关性以及作为全局特征提取方法可能会失去一些重要的局部特征等问题,提出一种基于块双向二维Fisher线性判别分析(B2DFLD)算法。首先利用块图像获取保持重要局部信息;然后基于行列双向投影,获取提取特征信息;最后计算特征阵的Frobenius距离,并进行分类。在ORL、YALE与FERET人脸数据库上进行了实验,并同传统的8种人脸识别方法比较。实验结果表明,在确定图像块大小、改变训练样本数以及特征维数的前提下,本文方法的最好识别率都高于93.08,平均误识率高于0.15,明显优于其他方法,表明本文方法对有光照、表情以及遮挡的人脸图像识别具有较高的鲁棒性。     

局部线性判别分析及其在掌纹识别中的应用

在线性判别分析(LDA)算法基础上强调图像类间数据的局部可分性,提出一种称为局部LDA(LLDA)的新子空间方法,并给出LLDA的图嵌入表示。针对LLDA同样存在的小样本问题,首先给出了传统适于LDA的主成分分析(PCA)预处理方法不适于LLDA算法的证明;进而提出了基于散度差判别准则(SDDC)的LLDA(SLLDA),既克服了LLDA的小样本问题,又提供了真实比较LLDA和LDA的平台。在PolyU掌纹数据库上的实验结果表明本文提出的SLLDA算法用于识别的有效性,也验证了数据局部关系对分类的重要性。     

基于核主元分析和Fisher线性判别的掌纹识别

提出了基于核主元分析(KPCA)和FLD相结合的掌纹识别方法.对每幅掌纹图像应用KPCA进行降维,然后将二维图像矩阵转换成一维图像矢量.PolyU掌纹图像库中所有图像矢量组成的数据矩阵作为FLD的输入,进行特征提取,计算特征矢量间的余弦距离进行掌纹匹配.实验结果说明,与传统的PCA+FLD相比,在不同的特征个数下,本文方法均取得了较小的等错率(EER),而且特征提取时间较短,运行速度较快.在三种不同的核函数中,RBF核函数的识别效果最佳,等错率最小为0.     

基于正则化零空间鉴别分析的人脸识别

提取有效的特征对高维数据的模式分类起着关键 作用,针对现有故障特征维数过高的问题,本文提 出了一种基于正则化零空间线性鉴别分析(Exponential Regularized Null Space Linear Discriminant Analysis, ERNSLDA)的特征提取方法。零空间线性判别分析已经在数据降维和特征提取上展现出良好 的性能,在 本文中,首先对类内样本矩阵进行正则化处理,避免小样本问题,其次对判别准则进行指数 化处理。所提 方法集成了NSLDA和RLDA在模式识别上的优势,有效地提高了人脸识别的精度,在ORL和YALE 数据库上的仿真实验证了本文所提方法的有效性。     

Multilinear discriminant analysis for face recognition.

There is a growing interest in subspace learning techniques for face recognition; however, the excessive dimension of the data space often brings the algorithms into the curse of dimensionality dilemma. In this paper, we present a novel approach to solve the supervised dimensionality reduction problem by encoding an image object as a general tensor of second or even higher order. First, we propose a discriminant tensor criterion, whereby multiple interrelated lower dimensional discriminative subspaces are derived for feature extraction. Then, a novel approach, called k-mode optimization, is presented to iteratively learn these subspaces by unfolding the tensor along different tensor directions. We call this algorithm multilinear discriminant analysis (MDA), which has the following characteristics: 1) multiple interrelated subspaces can collaborate to discriminate different classes, 2) for classification problems involving higher order tensors, the MDA algorithm can avoid the curse of dimensionality dilemma and alleviate the small sample size problem, and 3) the computational cost in the learning stage is reduced to a large extent owing to the reduced data dimensions in k-mode optimization. We provide extensive experiments on ORL, CMU PIE, and FERET databases by encoding face images as second- or third-order tensors to demonstrate that the proposed MDA algorithm based on higher order tensors has the potential to outperform the traditional vector-based subspace learning algorithms, especially in the cases with small sample sizes.     

Fault detection and isolation using concatenated wavelet transform variances and discriminant analysis

A method for fault detection and isolation is developed using the concatenated variances of the continuous wavelet transform (CWT) of plant outputs. These concatenated variances are projected onto the principal component space corresponding to the covariance matrix of the concatenated variances. Fisher and quadratic discriminant analyses are then performed in this space to classify the concatenated sample CWT variances of outputs in a given time window. The sample variance is a variance estimator obtained by taking the displacement average of the squared wavelet transforms of the current outputs. This method provides an alternative to the multimodel approach used for fault detection and identification, especially when system inputs are unmeasured stochastic processes, as is assumed in the case of the mechanical system example. The performance of the method is assessed using matrices having the percentage of correct condition identification in the diagonal and the percentages misclassified conditions in the off-diagonal elements. Considerable performance improvements may be obtained due to concatenation-when two or more outputs are available-and to discriminant analysis, as compared with other wavelet variance methods.     

基于直接辨别分析的雷达目标一维距离像识别

提出了基于零空间的线性直接辨别分析与非线性推广直接辨别分析方法,并将其用于雷达目标一维距离像识别.与传统子空间方法相比,上述两种方法保留并充分利用了类内散度矩阵最具分辨力的零空间信息,因而大大提高了目标的识别性能.对三种实测飞机数据的识别结果表明了所提方法的有效性.     

Human action recognition using Pose-based discriminant embedding

Manifold learning is an efficient approach for recognizing human actions. Most of the previous embedding methods are learned based on the distances between frames as data points. Thus they may be efficient in the frame recognition framework, but they will not guarantee to give optimum results when sequences are to be classified as in the case of action recognition in which temporal constraints convey important information. In the sequence recognition framework, sequences are compared based on the distances defined between sets of points. Among them Spatio-temporal Correlation Distance (SCD) is an efficient measure for comparing ordered sequences. In this paper we propose a novel embedding which is optimum in the sequence recognition framework based on SCD as the distance measure. Specifically, the proposed embedding minimizes the sum of the distances between intra-class sequences while seeking to maximize the sum of distances between inter-class points. Action sequences are represented by key poses chosen equidistantly from one action period. The action period is computed by a modified correlation-based method. Action recognition is achieved by comparing the projected sequences in the low-dimensional subspace using SCD or Hausdorff distance in a nearest neighbor framework. Several experiments are carried out on three popular datasets. The method is shown not only to classify the actions efficiently obtaining results comparable to the state of the art on all datasets, but also to be robust to additive noise and tolerant to occlusion, deformation and change in view point. Moreover, the method outperforms other classical dimension reduction techniques and performs faster by choosing less number of postures.     

Gait recognition using Pose Kinematics and Pose Energy Image

Many of the existing gait recognition approaches represent a gait cycle using a single 2D image called Gait Energy Image (GEI) or its variants. Since these methods suffer from lack of dynamic information, we model a gait cycle using a chain of key poses and extract a novel feature called Pose Energy Image (PEI). PEI is the average image of all the silhouettes in a key pose state of a gait cycle. By increasing the resolution of gait representation, more detailed dynamic information can be captured. However, processing speed and space requirement are higher for PEI than the conventional GEI methods. To overcome this shortcoming, another novel feature named as Pose Kinematics is introduced, which represents the percentage of time spent in each key pose state over a gait cycle. Although the Pose Kinematics based method is fast, its accuracy is not very high. A hierarchical method for combining these two features is, therefore, proposed. At first, Pose Kinematics is applied to select a set of most probable classes. Then, PEI is used on these selected classes to get the final classification. Experimental results on CMU's Mobo and USF's HumanID data set show that the proposed approach outperforms existing approaches.     

基于二维Fisher线性判别的人脸耳组合识别

针对人脸易受到年龄、表情等影响,提出了脸和耳相结合的组合识别方法。利用二维Fisher线性判别(2DFLD)方法分别进行了脸、耳图像层和特征层的组合识别。在北京科技大学人耳库和ORL人脸库上进行实验,结果表明,图像层组合和特征层组合的识别率分别为97.5%、95.0%,分别比人脸识别提高了12.5%和10.0%,比人耳识别提高了5.0%和2.5%;与同样应用于组合识别的主成分分析(PCA)、二维PCA(2DPCA)比较,也取得了较好识别效果。这说明,多生物特征组合识别是一种有效的识别方法。     

Optical computation of the sector and radon transform using pinhole array

An optical setup for the computation of the sector transform is presented in this paper. The device is based on pinhole imaging and processing with the computer generated mask. Computation of the transform does not require image rotation, and therefore processing is done fully in parallel. Extensions of the transform are also discussed.     

核Fisher鉴别分析在掌纹识别中的应用

核Fisher的鉴别方法(KFDA)是模式识别中较为突出的提取图像非线性特征的方法。为了更好的提取掌纹图像的非线性特征,将KFDA方法引入到掌纹识别中。首先对掌纹图像做小波变换进行降维,在保留原始图像轮廓信息和特征的基础上,然后进行核Fisher判决方法进行特征提取并引入零空间的核Fisher(ZKFDA)方法解决小样本问题,最后用最小距离分类器进行掌纹匹配。通过PolyU掌纹图像库,实验结果表明,在不同的特征个数下,KFDA方法比二维Fisher准则(2DFLD)方法识别率高;零空间ZKFDA的平均识别率高于KFDA,并且计算量大大减少。在核函数选取上,取RBF核函数的识别性能最佳。     

基于无穷范数的二值线性判别分析

线性判别分析（LDA）是监督式的特征提取方法,在人脸识别等领域得到了广泛应用。为了提高特征提取速度,提出了基于无穷范数的线性判别分析方法。传统LDA方法将目标函数表示为类内散布矩阵和类间散布矩阵之差的或者商的L2范数,且通常需要涉及到矩阵求逆和特征值分解问题。与传统方法不同,这里所提方法将目标函数表示为类内散布矩阵和类间散布矩阵之差的无穷范数,而且最优解是以迭代形式得到,避免了耗时的特征值分解。无穷范数使得到的基向量实现了二值化,即元素仅在-1和1两个数字内取值,避免了特征提取时的浮点型点积运算,从而降低了测试时间,提高了效率。在ORL人脸数据库和Yale数据库上的实验表明所提算法是有效的。     

Accurate centerline detection and line width estimation of thick lines using the radon transform.

Centerline detection and line width estimation are important for many computer vision applications, e.g., road network extraction from high resolution remotely sensed imagery. Radon transform-based linear feature detection has many advantages over other approaches: for example, its robustness in noisy images. However, it usually fails to detect the centerline of a thick line due to the peak selection problem. In this paper, several key issues that affect the centerline detection using the radon transform are investigated. A mean filter is proposed to locate the true peak in the radon image and a profile analysis technique is used to further refine the line parameters. The theta-boundary problem of the radon transform is also discussed and the erroneous line parameters are corrected. Intensive experiments have shown that the proposed methodology is effective in finding the centerline and estimating the line width of thick lines.     

Rotation-invariant multiresolution texture analysis using radon and wavelet transforms.

A new rotation-invariant texture-analysis technique using Radon and wavelet transforms is proposed. This technique utilizes the Radon transform to convert the rotation to translation and then applies a translation-invariant wavelet transform to the result to extract texture features. A kappa-nearest neighbors classifier is employed to classify texture patterns. A method to find the optimal number of projections for the Radon transform is proposed. It is shown that the extracted features generate an efficient orthogonal feature space. It is also shown that the proposed features extract both of the local and directional information of the texture patterns. The proposed method is robust to additive white noise as a result of summing pixel values to generate projections in the Radon transform step. To test and evaluate the method, we employed several sets of textures along with different wavelet bases. Experimental results show the superiority of the proposed method and its robustness to additive white noise in comparison with some recent texture-analysis methods.