A sparse representation method based on kernel and virtual samples for face recognition

To improve the classification accuracy of face recognition, a sparse representation method based on kernel and virtual samples is proposed in this paper. The proposed method has the following basic idea: first, it extends the training samples by copying the left side of the original training samples to the right side to form virtual training samples. Then the virtual training samples and the original training samples make up a new training set and we use a kernel-induced distance to determine M nearest neighbors of the test sample from the new training set. Second, it expresses the test sample as a linear combination of the selected M nearest training samples and finally exploits the determined linear combination to perform classification of the test sample. A large number of face recognition experiments on different face databases illustrate that the error ratios obtained by our method are always lower more or less than face recognition methods including the method mentioned in Xu and Zhu [21], the method proposed in Xu and Zhu [39], sparse representation method based on virtual samples (SRMVS), collaborative representation based classification with regularized least square (CRC_RLS), two-phase test sample sparse representation (TPTSSR), and the feature space-based representation method.     

A novel virtual samples-based sparse representation method for face recognition

A limited training set usually limits the performance of face recognition in practice. Even sparse representation-based methods which outperform in face recognition cannot avoid such situation. In order to effectively improve recognition accuracy of sparse representation-based methods on a limited training set, a novel virtual samples-based sparse representation (VSSR) method for face recognition is proposed in this paper. In the proposed method, virtual training samples are constructed to enrich the size and diversity of a training set and a sparse representation-based method is used to classify test samples. Extensive experiments on different face databases confirm that VSSR is robust to illumination variations and works better than many representative representation-based face recognition methods.     

Integrating the original face images and “symmetrical faces” to perform face recognition

Using the original and ‘symmetrical face’ training samples to perform representation based face recognition was first proposed in [1]. It simultaneously used the original and ‘symmetrical face’ training samples to perform a two-step classification and achieved an outstanding classification result. However, in [1] the “symmetrical face” is devised only for one method. In this paper, we do some improvements on the basis of [1] and combine this “symmetrical faces” transformation with several representation based methods. We exploit all original training samples, left “symmetrical face” training samples and right “symmetrical face” training samples for classification and use the score fusion for ultimate face recognition. The symmetry of the face is first used to generate new samples, which is different from original face image but can really reflect some possible appearance of the face. It effectively overcomes the problem of non-sufficient training samples. The experimental results show that the proposed scheme can be used to improve a number of traditional representation based methods including those that are not presented in the paper.     

Quaternion-based improved LPP method for color face recognition

In recent years, pattern recognition and computer vision have increasingly become the focus of research. Locality preserving projection (LPP) is a very important learning method in these two fields and has been widely used. Using LPP to perform face recognition, we usually can get a high accuracy. However, the face recognition application of LPP suffers from a number of problems and the small sample size is the most famous one. Moreover, though the face image is usually a color image, LPP cannot sufficiently exploit the color and we should first convert the color image into the gray image and then apply LPP to it. Transforming the color image into the gray image will cause a serious loss of image information. In this paper, we first use the quaternion to represent the color pixel. As a result, an original training or test sample can be denoted as a quaternion vector. Then we apply LPP to the quaternion vectors to perform feature extraction for the original training and test samples. The devised quaternion-based improved LPP method is presented in detail. Experimental results show that our method can get a higher classification accuracy than other methods.     

A two-phase face recognition method in frequency domain

In this paper, we propose a two-phase face recognition method in frequency domain using discrete cosine transform (DCT) and discrete Fourier transform (DFT). The absolute values of DCT coefficients or DFT amplitude spectra are used to represent the face image, i.e. the transformed image. Then a two-phase face classification method is applied to the transformed images. This method is as follows: its first phase uses the Euclidean distance formula to calculate the distance between a test sample and each sample in the training sets, and then exploits the Euclidean distance of each training sample to determine K nearest neighbors for the test sample. Its second phase represents the test sample as a linear combination of the determined K nearest neighbors and uses the representation result to perform classification. In addition, we use various numbers of DCT coefficients and DFT amplitude spectra to test the effect on our algorithms. The experimental results show that our method outperforms the two-phase face recognition method based on space domain of face images.     

A hierarchical structure with improved OMP sparse representation used with face recognition

With the rapid development of the face recognition technology, more and more optical products are applied in people's real life. The recognition accuracy can be improved by increasing the number of training samples, but the colossal training samples will result in the increase of computational complexity. In recent years, sparse representation method becomes a research hot spot on face recognition. In this paper we propose an energy constrain orthogonal matching pursuit (ECOMP) algorithm for sparse representation to select the few training samples and a hierarchical structure for face recognition. We filter the training samples with ECOMP algorithm and then we compute the weights by all selected training samples. At last we find the closest recovery sample to the test sample. Simultaneously the experimental results in AR, ORL and FERET database also show that our proposed method has better recognition performance than the LRC and SRC_OMP method.     

‘Symmetrical face’ based improved LPP method for face recognition

As one of the most important branches of pattern recognition and computer vision, face recognition has more and more become the focus of researches. In real word applications, the face image might have various changes owing to varying illumination, facial expression and poses, so we need sufficient training samples to convey these possible changes. However, most face recognition systems cannot capture many face images of every user for training, non-sufficient training samples have become one bottleneck of face recognition. In this paper, we propose to exploit the symmetry of the face to generate ‘symmetrical face’ samples and use an improved LPP method to perform classification. Experimental results show that our method can get a high accuracy.     

基于WSSRC单样本人脸识别及样本扩充方法研究

由于传统的SRC方法的实时性不强、单样本条件下算法性能低等缺点,提出了融合全局和局部特征的加权超级稀疏表示人脸识别方法(WSSRC),同时采用一种三层级联的虚拟样本产生方法获取冗余样本,将生成的多种表情和多种姿态的新样本当成训练样本,运用WSSRC算法进行人脸识别分类。在单样本的情况下,实验证实在ORL人脸库上该方法比传统的SRC方法提高了15.53%的识别率,使用在FERET 人脸库上则提高7.67%。这样的方法与RSRC 、SSRC、DMMA、DCT-based DMMA、I-DMMA相比,一样具备较好的识别性能。     

A novel pose and illumination robust face recognition with a single training image per person algorithm

In the real-world application of face recognition system, owing to the difficulties of collecting samples or storage space of systems, only one sample image per person is stored in the system, which is so-called one sample per person problem. Moreover, pose and illumination have impact on recognition performance. We propose a novel pose and illumination robust algorithm for face recognition with a single training image per person to solve the above limitations. Experimental results show that the proposed algorithm is an efficient and practical approach for face recognition.     

基于人脸姿态估计的虚拟眼镜试戴技术

为解决用户线上眼镜的最优选购,提出了一种基于人脸姿态估计的虚拟眼镜试戴技术。首先采用肤色模型与形状模型结合的算法对场景中的人脸区域进行检测,然后根据人眼在人脸中的几何位置关系实现人眼的精确定位;进一步利用人眼对称性先验知识来估计脸在三维空间中的姿态信息,即人脸与正面的偏移角度;最后,依据人眼位置和人脸姿态将眼镜图像融合到眼睛区域,即完成眼镜的虚拟试戴。该方法为3D环境下客户与商品之间虚拟视觉化的实现提供了一种可靠的技术支撑和应用思路。     

Feature fusion of palmprint and face via tensor analysis and curvelet transform

In order to improve the recognition accuracy of the unimodal biometric system and to address the problem of the small samples recognition, a multimodal biometric recognition approach based on feature fusion level and curve tensor is proposed in this paper. The curve tensor approach is an extension of the tensor analysis method based on curvelet coefficients space. We use two kinds of biometrics: palmprint recognition and face recognition. All image features are extracted by using the curve tensor algorithm and then the normalized features are combined at the feature fusion level by using several fusion strategies. The k-nearest neighbour (KNN) classifier is used to determine the final biometric classification. The experimental results demonstrate that the proposed approach outperforms the unimodal solution and the proposed nearly Gaussian fusion (NGF) strategy has a better performance than other fusion rules.     

Face recognition based on PCA image reconstruction and LDA

Face recognition has become a research hotspot in the field of pattern recognition and artificial intelligence. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) are two traditional methods in pattern recognition. In this paper, we propose a novel method based on PCA image reconstruction and LDA for face recognition. First, the inner-classes covariance matrix for feature extraction is used as generating matrix and then eigenvectors from each person is obtained, then we obtain the reconstructed images. Moreover, the residual images are computed by subtracting reconstructed images from original face images. Furthermore, the residual images are applied by LDA to obtain the coefficient matrices. Finally, the features are utilized to train and test SVMs for face recognition. The simulation experiments illustrate the effectivity of this method on the ORL face database.     

Modular Fisher Discriminant Sparse Representation for robust face recognition

This paper proposes a novel framework for robust face recognition based on sparse representation and discrimination ranking. This method consists of three stages. The first stage partitions each training sample into some overlapped modules and then computes each module's Fisher ratio, respectively. The second stage selects modules which have higher Fisher ratios to comprise a template to filter training and test images. The dictionary is constructed by the filtered training images. The third stage computes the sparse representation of filtered test sample on the dictionary to perform identification. The advantages of the proposed method are listed as follows: the first stage can preserve the local structure. The second stage removes the modules that have little contribution for classification. Then the method uses the retaining modules to classify the test sample by SRC which makes the method robust. Compared with the related methods, experimental results on benchmark face databases verify the advancement of the proposed method. The proposed method not only has a high accuracy but also can be clearly interpreted.     

Kernel-aligned multi-view canonical correlation analysis for image recognition

Existing kernel-based correlation analysis methods mainly adopt a single kernel in each view. However, only a single kernel is usually insufficient to characterize nonlinear distribution information of a view. To solve the problem, we transform each original feature vector into a 2-dimensional feature matrix by means of kernel alignment, and then propose a novel kernel-aligned multi-view canonical correlation analysis (KAMCCA) method on the basis of the feature matrices. Our proposed method can simultaneously employ multiple kernels to better capture the nonlinear distribution information of each view, so that correlation features learned by KAMCCA can have well discriminating power in real-world image recognition. Extensive experiments are designed on five real-world image datasets, including NIR face images, thermal face images, visible face images, handwritten digit images, and object images. Promising experimental results on the datasets have manifested the effectiveness of our proposed method.     

A face recognition algorithm based on collaborative representation

In this paper, we propose a face recognition algorithm by incorporating a neighbor matrix into the objective function of sparse coding. We first calculate the neighbor matrix between the test sample and each training sample by using the revised reconstruction error of each class. Specifically, the revised reconstruction error (RRE) of each class is the division of the l₂-norm of reconstruction error to the l₂-norm of reconstruction coefficients, which can be used to increase the discrimination information for classification. Then we use the neighbor matrix and all the training samples to linearly represent the test sample. Thus, our algorithm can preserve locality and similarity information of sparse coding. The experimental results show that our algorithm achieves better performance than four previous algorithms on three face databases.     

基于子空间分析的人脸识别方法研究

人脸识别技术是模式识别和机器视觉领域的一个重要研究方向,在众多人脸识别的算法中,基于子空间分析的特征提取方法以其稳定可靠的识别效果成为了人脸识别中特征提取的主流方法之一。本文对目前应用较多的子空间分析方法进行了研究,具体介绍了线性子空间分析方法：主成分分析（PCA）、线性鉴别分析（LDA）、独立主成分分析（ICA）、快速主成分分析（FastICA）等及非线性子空间分析方法：基于核的PCA （KPCA）等的基本思想及其在人脸识别中的研究进展,包括一些新的研究成果。此外,还应用orl及Yale B人脸库对几个基础的子空间方法进行了验证实验。实验结果表明,在几个子空间分析方法中,FastICA算法取得了最高的识别率。最后结合实验结果对各算法的优缺点进行了分析总结。     

Within-component and between-component discriminant analysis for color face recognition

How to efficiently utilize the color image information and extract effective features is the key of color face recognition. In this paper, we first analyze the similarities between facial color component image samples and their influence on color face recognition. Then we propose a novel color face recognition approach named within-component and between-component discriminant analysis (WBDA), which realizes discriminant analysis not only within each color component but also between different components. Experimental results on the face recognition grand challenge (FRGC) version 2 database demonstrate that the proposed approach outperforms several representative color face recognition methods.     

Double fusion filtering based multi-view face recognition

Two-dimensional (2D) face recognition by correlation is a key challenge of telecommunication and optical information processing. Although this issue has been the focus of intense research, its utilization still has some drawbacks especially when the face is in rotation. In this paper, we propose an alternative method based on a newly designed optical correlation filter which allows recognizing faces with different view angles. This filter called “Multi-View Binary Phase-Only Filter” is based on a double fusion of reference images allowing an optimisation of the use of the spatial-bandwidth product (SBWP) in the filter Fourier plane. The first fusion is performed in the image (space) domain, and the second one is conducted in the spectral domain. Simulations results with the Pointing Head Pose Image Database illustrate the performance of the designed correlation filter for multi-view face recognition.     

Super-resolution reconstruction face recognition based on multi-level FFD registration

Super-resolution (SR) reconstruction is an effective method to solve the problem, that the face image resolution is too low to be recognized in video, but the non-rigid change of deformed face and expression changes greatly affect the accuracy of registration and reconstruction. To solve these problems, a method of multi-level model free form deformation (FFD) elastic registration algorithm based on B spline is proposed. It first use low-resolution FFD grid for global registration, to emphasize the contribution of edge information for registration, we introduce edge registration measure into the sum of squared difference (SSD) criterion. Then divide the global registration image and reference image into a series of corresponding sub-image pairs and calculate the correlation coefficient of each pair; at the same time, we do local registration with high-resolution FFD grid to the small value correlation coefficient sub-image pairs. In the registration process of optimization, the paper uses adaptive step length gradient descent method algorithm based on chaotic variables to improve optimization efficiency. After registration, the algorithm of project onto convex sets (POCS) is used to reconstruct SR face image through several low resolution image sequences, and then recognized these SR face images by support vector machines (SVM) classifier. Experimental results from standard video database and self-built video database show that this method can register and reconstruct face image accurately in the condition of great face deformation and expression change, while the face recognition accuracy is also improved.