Color space normalization: Enhancing the discriminating power of color spaces for face recognition

This paper presents the concept of color space normalization (CSN) and two CSN techniques, i.e., the within-color-component normalization technique (CSN-I) and the across-color-component normalization technique (CSN-II), for enhancing the discriminating power of color spaces for face recognition. Different color spaces usually display different discriminating power, and our experiments on a large scale face recognition grand challenge (FRGC) problem reveal that the RGB and XYZ color spaces are weaker than the I₁I₂I₃, YUV, YIQ, and LSLM color spaces for face recognition. We therefore apply our CSN techniques to normalize the weak color spaces, such as the RGB and the XYZ color spaces, the three hybrid color spaces XGB, YRB and ZRG, and 10 randomly generated color spaces. Experiments using the most challenging FRGC version 2 Experiment 4 with 12,776 training images, 16,028 controlled target images, and 8,014 uncontrolled query images, show that the proposed CSN techniques can significantly and consistently improve the discriminating power of the weak color spaces. Specifically, the normalized RGB, XYZ, XGB, and ZRG color spaces are more effective than or as effective as the I₁I₂I₃, YUV, YIQ and LSLM color spaces for face recognition. The additional experiments using the AR database validate the generalization of the proposed CSN techniques. We finally explain why the CSN techniques can improve the recognition performance of color spaces from the color component correlation point of view.     

Learning the Uncorrelated, Independent, and Discriminating Color Spaces for Face Recognition

This paper presents learning the uncorrelated color space (UCS), the independent color space (ICS), and the discriminating color space (DCS) for face recognition. The new color spaces are derived from the RGB color space that defines the tristimuli R, G, and B component images. While the UCS decorrelates its three component images using principal component analysis (PCA), the ICS derives three independent component images by means of blind source separation, such as independent component analysis (ICA). The DCS, which applies discriminant analysis, defines three new component images that are effective for face recognition. Effective color image representation is formed in these color spaces by concatenating their component images, and efficient color image classification is achieved using the effective color image representation and an enhanced Fisher model (EFM). Experiments on the face recognition grand challenge (FRGC) and the biometric experimentation environment (BEE) show that for the most challenging FRGC version 2 Experiment 4, which contains 12 776 training images, 16 028 controlled target images, and 8014 uncontrolled query images, the ICS, DCS, and UCS achieve the face verification rate (ROC III) of 73.69%, 71.42%, and 69.92%, respectively, at the false accept rate of 0.1%, compared to the RGB color space, the 2-D Karhunen-Loeve (KL) color space, and the FRGC baseline algorithm with the face verification rate of 67.13%, 59.16%, and 11.86%, respectively, with the same false accept rate.     

Color Image Discriminant Models and Algorithms for Face Recognition

This paper presents a basic color image discriminant (CID) model and its general version for color image recognition. The CID models seek to unify the color image representation and recognition tasks into one framework. The proposed models, therefore, involve two sets of variables: a set of color component combination coefficients for color image representation and one or multiple projection basis vectors for color image discrimination. An iterative basic CID algorithm and its general version are designed to find the optimal solution of the proposed models. The general CID (GCID) algorithm is further extended to generate three color components (such as the three color components of the RGB color images) for further improvement of the recognition performance. Experiments using the face recognition grand challenge (FRGC) database and the biometric experimentation environment (BEE) system show the effectiveness of the proposed models and algorithms. In particular, for the most challenging FRGC version 2 Experiment 4, which contains 12 776 training images, 16 028 controlled target images, and 8014 uncontrolled query images, the proposed method achieves the face verification rate (ROC III) of 78.26% at the false accept rate (FAR) of 0.1%.      

Fusion of local normalization and Gabor entropy weighted features for face identification

Face recognition is one of the most extensively studied topics in image analysis because of its wide range of possible applications such as in surveillance, access control, content-based video search, human–computer interaction, electronic advertisement and more. Face identification is a one-to-n matching problem where a captured face is compared to n samples in a database. In this work we propose two new methods for face identification. The first one combines entropy-like weighted Gabor features with the local normalization of Gabor features. The second fuses the entropy-like weighted Gabor features at the score level with the local binary pattern (LBP) applied to the magnitude (LGBP) and phase (LGXP) components of the Gabor features. We used the FERET, AR, and FRGC 2.0 databases to test and compare our results with those previously published. Results on these databases show significant improvement relative to previously published results, reaching the best performance on the FERET and AR databases. Our methods also showed significant robustness to slight pose variations. We tested the proposed methods assuming noisy eye detection to check their robustness to inexact face alignment. Results show that the proposed methods are robust to errors of up to 3 pixels in eye detection.     

Computing (r, Q) policy for an inventory system with limited sharable resource

This paper deals with inventory systems with limited resource for a single item or multiple items under continuous review (r, Q) policies. For the single-item system with a stochastic demand and limited resource, it is shown that an existing algorithm can be applied to find an optimal (r, Q) policy that minimizes the expected system costs. For the multi-item system with stochastic demands and limited resource commonly shared among all items, an optimization problem is formulated for finding optimal (r, Q) policies for all items, which minimize the expected system costs. Bounds on the parameters (i.e., r and Q) of the optimal policies and bounds on the minimum expected system costs are obtained. Based on the bounds, an algorithm is developed for finding an optimal or near-optimal solution. A method is proposed for evaluating the quality of the solution. It is shown that the algorithm proposed in this paper finds a solution that is (i) optimal/near-optimal and/or (ii) significantly better than the optimal solution with unlimited resource.     

Extracting discriminative color features for face recognition

This paper presents a discriminative color features (DCF) method, which applies a simple yet effective color model, a novel similarity measure, and effective color feature extraction methods, for improving face recognition performance. First, the new color model is constructed according to the principle of Ockham’s razor from a number of available models that take advantage of the subtraction of the primary colors for boosting pattern recognition performance. Second, the novel similarity measure integrates both the angular and the distance information for improving upon the broadly applied similarity measures. Finally, the discriminative color features are extracted from a compact color image representation by means of discriminant analysis with enhanced generalization capabilities. Experiments on the Face Recognition Grand Challenge (FRGC) version 2 Experiment 4, which contains 12,776 training images, 16,028 controlled target images, and 8,014 uncontrolled query images, show the feasibility of the proposed method.     

基于核典型相关分析的多特征组合人脸识别

为了提高人脸的识别率,利用多特征和分类器之间的互补优势,提出一种基于核典型相关分析的多特征组合人脸识别方法（KCCA-MF）。提取人脸图像的LBP特征和Gabor特征,采用核典型相关分析算法对两种特征进行融合,以消除冗余特征,采用K近邻算法和支持向量机建立组合人脸分类器,并采用3个经典人脸库进行仿真分析。结果表明,相对于其他人脸识别方法,KCCA-MF提高了人脸识别的识别准确率和效率,可以满足人脸识别的实时性要求。     

基于局部Gabor变化直方图序列的人脸描述与识别

提出了一种在Gabor变换幅值域内提取局部变化模式空间直方图序列(histogram sequence of local Gabor binary patterns,简称HSLGBP)的人脸描述及其识别方法.鉴于Gabor特征对光照、表情等变化比较鲁棒,并已在人脸识别领域得到成功应用,首先对归一化的人脸图像进行多方向、多分辨率Gabor小波滤波,并提取其对应不同方向、不同尺度的多个Gabor幅值域图谱(Gabor magnitude map,简称GMM),然后在每个GMM上采用局部二值模式(local binary pattern,简称LBP)算子抽取局部邻域关系模式,最后由这些模式的区域直方图形成的序列来描述人脸.Gabor变换、LBP、空间区域直方图的采用使得该方法对光照变化、表情变化、误配准等具有良好的鲁棒性.而且,这种人脸建模方法不需要基于训练集合进行统计学习,因而不存在推广性问题.同时,进一步探讨了如何在分类器设计阶段与统计方法进行结合的问题,提出了统计Fisher加权的HSLGBP匹配方法.在通过FERET人脸库光照、表情和时间变化测试集上与已发表的实验结果进行对比,充分验证了该方法的有效性.     

Analysis of variance of Gabor filter banks parameters for optimal face recognition

Gabor filter banks constitute a very robust tool to extract discriminant information from a visual scene. After the now “classical” bank with 5 frequencies and 8 orientations proposed by Lades et al. and Wiskott et al., many other parametrizations of a Gabor filter bank have appeared. In order to find the optimal parametrization for a face recognition experiment, we have performed a 6-way analysis of variance of Gabor parameters using FERET, FRAV2D, FRAV3D, FRGC and XM2VTS face databases, including frontal and turned poses, facial expressions, occlusions and changes of illumination. Considering independent criteria to find the optimal Gabor filter bank, the bank with the highest recognition rate was found to have 6 frequencies and narrower Gaussian widths in the space domain. These results were obtained with Mahalanobis distance for a k-NN classifier, with analytical and holistic Gabor feature vectors. Moreover about 20% of the banks studied here obtained in average a better performance than the classical bank. For most of the databases considered, the highest recognition rates have been achieved with analytical representations (frontal images, images with turns or occlusions), with a holistic preponderance for images with gestures or changes of illumination. The inferiority found for holistic Gabor representations versus their analytical counterparts can be explained for the intrinsic redundancy and the size of the feature vectors of this kind of representation.     

基于四元数Gabor滤波的彩色纹理特征提取

现有的彩色图像纹理特征提取方法是将彩色图像转换为灰度图像或者对彩色图像进行分通道处理,这样的处理方法会丢失原图像的颜色信息和各通道间的相关性,导致特征图像的纹理特征和原图像的纹理特征差异较大。基于上述问题,提出了一种四元数Gabor彩色纹理特征提取方法。首先,根据Gabor滤波和四元数欧拉公式,推导出四元数Gabor滤波,并将彩色图像用四元数矩阵表达;其次提出四元数Gabor滤波卷积算法处理彩色图像,得到多尺度多方向的彩色纹理特征图像;最后对得到的彩色纹理特征图像进行Tamura统计特征的提取。实验结果表明,该方法可以很大程度地保留原图像的粗糙度、对比度和方向度等纹理特征,同时可以提取到原图像的颜色信息。在转化为灰度图像后,该方法在保留粗糙度、对比度和方向度等纹理特征方面优于传统Gabor方法和LBP方法。     

Capitalize on dimensionality increasing techniques for improving Face Recognition Grand Challenge performance

This paper presents a novel pattern recognition framework by capitalizing on dimensionality increasing techniques. In particular, the framework integrates Gabor image representation, a novel multiclass Kernel Fisher Analysis (KFA) method, and fractional power polynomial models for improving pattern recognition performance. Gabor image representation, which increases dimensionality by incorporating Gabor filters with different scales and orientations, is characterized by spatial frequency, spatial locality, and orientational selectivity for coping with image variabilities such as illumination variations. The KFA method first performs nonlinear mapping from the input space to a high-dimensional feature space, and then implements the multiclass Fisher discriminant analysis in the feature space. The significance of the nonlinear mapping is that it increases the discriminating power of the KFA method, which is linear in the feature space but nonlinear in the input space. The novelty of the KFA method comes from the fact that 1) it extends the two-class kernel Fisher methods by addressing multiclass pattern classification problems and 2) it improves upon the traditional Generalized Discriminant Analysis (GDA) method by deriving a unique solution (compared to the GDA solution, which is not unique). The fractional power polynomial models further improve performance of the proposed pattern recognition framework. Experiments on face recognition using both the FERET database and the FRGC (Face Recognition Grand Challenge) databases show the feasibility of the proposed framework. In particular, experimental results using the FERET database show that the KFA method performs better than the GDA method and the fractional power polynomial models help both the KFA method and the GDA method improve their face recognition performance. Experimental results using the FRGC databases show that the proposed pattern recognition framework improves face recognition performance upon the BEE baseline algorithm and the LDA-based baseline algorithm by large margins.     

Perfect r-domination in the Kronecker product of two cycles, with an application to diagonal/toroidal mesh

If r?1, and m and n are each a multiple of (r+1)²+r², then each isomorphic component of C_m×C_n admits of a vertex partition into (r+1)²+r² perfect r-dominating sets. The result induces a dense packing of C_m×C_n by means of vertex-disjoint subgraphs, each isomorphic to a diagonal array. Areas of applications include efficient resource placement in a diagonal mesh and error-correcting codes.     

基于LBP和Fisherfaces的多模态人脸识别

提出一种结合局部二值模式（LBP）和Fisherfaces的多模态人脸识别方法。用LBP算子提取人脸灰度图像和深度图像的区域LBP直方图序列（LBPHS）,再采用Fisherfaces分别构建相应的线性子空间,用余弦相似度作为投影向量的相似度量,用加权求和规则进行信息融合。在FRGC数据库上的实验结果表明,该方法要明显优于LBPHS与直方图交及Fisherfaces与余弦相似度的融合,等错误率仅为0．33％。     

LBP与鉴别模式结合的热红外人脸识别

为了提取具有鉴别能力的红外人脸图像局部结构特征,提出一种基于LBP(local binary pattern)鉴别模式的红外人脸识别方法。传统的LBP均匀模式,提取自然图像中占主导地位的信息用于识别,但占主导地位的信息不一定是最适合识别的。为了提取有效的鉴别模式特征,基于监督学习的思想,在LBP模式下引入可分性标准,对不同LBP模式进行有效的模式选择,从而抽取适合识别的鉴别模式。最后,为了利用人脸的空间位置信息,结合分块和直方图技术得到最后的识别特征。实验结果表明,本文鉴别模式可以提取更适合识别的特征,识别性能优于传统的基于均匀模式的LBP方法。     

融合DCT和LBP特征的表情识别

为了获得更好的面部表情特征,提出了一种融合离散余弦变换（Discrete Cosine Transform,DCT）特征和局部二值模式（Local Binary Pattern,LBP）特征的表情特征提取方法。该方法将人脸图像经过DCT后所获得的低频系数作为表情的整体特征;通过对人脸图像进行分块,计算每个子块的LBP直方图,将这些LBP直方图连接起来形成LBP特征,对该LBP特征使用拉普拉斯特征映射（Laplacian Eigenmaps,LE）降维后得到表情的局部特征。将得到的整体特征和局部特征进行加权融合,使用最近邻分类器进行分类。在JAFFE和Cohn-Kanade表情库上的实验结果表明,该方法比单独使用LBP或者DCT特征,具有更好的效果。     

彩色图像统计正交分析方法

彩色图像含有比灰度图像更丰富的信息,因此在图像识别中扮演重要的角色。RGB彩色空间是使用最为广泛的彩色空间。通常R、G、B三分量间存在相关性。彩色图像识别技术的关键是如何有效使用分量间的补信息、消除冗余,并且提取有效的鉴别特征。文中提出了一种新的彩色图像特征提取方法,即彩色图像统计正交分析（CISOA）。该方法按照R、G、B的顺序依次提取三分量的鉴别特征,并保证各分量所提取的特征满足统计正交约束。在彩色人脸和掌纹图像数据库的实验结果表明此方法具有较好的识别效果。     

Pricing and inventory management in a system with multiple competing retailers under (r, Q) policies

We consider a multi-retailer system operated on an infinite horizon, in which each retailer faces stochastic demand following a Poisson process and adopts a continuous-review (r, Q) policy for replenishing inventory to satisfy customer demand. The system involves decisions of pricing and inventory management with the goal of maximizing profit, which equals the sales revenue minus the purchase and inventory costs. Taking Cournot competition into account, models are formulated to optimize simultaneously the expected sales volumes and (r, Q) policies of all retailers. An efficient approach is proposed to calculate the approximate inventory cost. Based on this approach, solution methods for centralized and decentralized scenarios are developed. A great number of numerical computations are provided to evaluate the efficiency of the solution methods, and their performance in the two scenarios. Moreover, system performance under sequential decisions (first pricing and then inventory management) is also investigated.     

Discriminant analysis and similarity measure

The cosine similarity measure is often applied after discriminant analysis in pattern recognition. This paper first analyzes why the cosine similarity is preferred by establishing the connection between the cosine similarity based decision rule in the discriminant analysis framework and the Bayes decision rule for minimum error. The paper then investigates the challenges inherent of the cosine similarity and presents a new similarity that overcomes these challenges. The contributions of the paper are thus three-fold. First, the application of the cosine similarity after discriminant analysis is discovered to have its theoretical roots in the Bayes decision rule. Second, some inherent problems of the cosine similarity such as its inadequacy in addressing distance and angular measures are discussed. Finally, a new similarity measure, which overcomes the problems by integrating the absolute value of the angular measure and the l_p norm (the distance measure), is presented to enhance pattern recognition performance. The effectiveness of the proposed new similarity measure in the discriminant analysis framework is evaluated using a large scale, grand challenge problem, namely, the Face Recognition Grand Challenge (FRGC) problem. Experimental results using 36,818 FRGC images on the most challenging FRGC experiment, the FRGC Experiment 4, show that the new similarity measure improves face recognition performance upon other popular similarity measures, such as the cosine similarity measure, the normalized correlation, and the Euclidean distance measure.     

基于火焰彩色纹理复杂度特征的转炉炼钢吹炼状态识别

在基于火焰图像识别的转炉吹炼状态识别过程中,针对已有方法存在火焰彩色纹理信息利用不充分和状态识别率仍需提高的问题,提出一种基于火焰彩色纹理复杂度特征的转炉吹炼状态识别方法.首先,将火焰图像转化到HSI颜色空间下并作非均匀量化;然后,计算H分量和S分量的共生矩阵从而融入火焰图像的颜色信息;其次,利用得到的颜色共生矩阵计算火焰纹理复杂度的特征描述子;最后,应用Canberra距离作为相似度度量准则对吹炼状态进行分类和识别.实验结果表明,与已有的转炉火焰灰度共生矩阵和灰度差分统计方法相比,在满足吹炼识别实时性要求的前提下,所提方法的识别率分别提高了28.33%和3.33%.     

Hybrid Human Skin Detection Using Neural Network and K-Means Clustering Technique

Human skin detection is an essential step in most human detection applications, such as face detection. The performance of any skin detection system depends on assessment of two components: feature extraction and detection method. Skin color is a robust cue used for human skin detection. However, the performance of color-based detection methods is constrained by the overlapping color spaces of skin and non-skin pixels. To increase the accuracy of skin detection, texture features can be exploited as additional cues. In this paper, we propose a hybrid skin detection method based on YIQ color space and the statistical features of skin. A Multilayer Perceptron artificial neural network, which is a universal classifier, is combined with the k-means clustering method to accurately detect skin. The experimental results show that the proposed method can achieve high accuracy with an F₁-measure of 87.82% based on images from the ECU database.