Combined features for face recognition under illumination variation

Face recognition has been widely studied due to its importance in different applications; however, most of the proposed methods fail when face images are occluded or captured under illumination and pose variations. Recently several low-rank dictionary learning methods have been proposed and achieved promising results for noisy observations. While these methods are mostly developed for single-modality scenarios, recent studies demonstrated the advantages of feature fusion from multiple inputs. We propose a multi-modal structured low-rank dictionary learning method for robust face recognition, using raw pixels of face images and their illumination invariant representation. The proposed method learns robust and discriminative representations from contaminated face images, even if there are few training samples with large intra-class variations. Extensive experiments on different datasets validate the superior performance and robustness of our method to severe illumination variations and occlusion.

Face recognition is one of the most successful applications in biometric authentication. However, methods reported in the literature still suffer from some problems which prevent the further development in face recognition. This paper presents a novel robust method for face recognition under near infrared (NIR) lighting condition based on Extended Local Binary Pattern (ELBP), which solves the problems produced by variations of illumination rightly, since the NIR images are insensitive to variations of ambient lighting, and ELBP can extract adequate texture features form the NIR images. By combining the local feature vectors, a global feature vector is formed and as the global feature vectors extracted by ELBP operator often have very high dimensions, a classifier has been trained using the AdaBoost algorithm to select the most representative features for better performance and dimensionality reduction. Compared with the huge number of features produced by ELBP operator, only a small part of the features are selected in this paper, which saves much computation and time cost. The comparison with the results of classic algorithms proves the effectiveness of the proposed method.

This paper presents a robust face recognition algorithm by using transform domain-based multiple feature fusion and linear regression. Transform domain-based feature fusion can provide comprehensive face information for recognition, and decrease the effect of variations in illumination and pose. The holistic feature and local feature are extracted by discrete cosine transform and Gabor wavelet transform, respectively. Then the extracted holistic features and the local features are fused by weighted sum. The fused feature values are finally sent to linear regression classifier for recognition. The algorithm is evaluated on AR, ORL and Yale B face databases. Experiment results show that our proposed algorithm could be more robust than those single feature-based algorithms under pose and expression variations.

An improved collaborative representation based classification with regularized least square (CRC–RLS) method for robust face recognition

Face recognition (FR) is a practical application of pattern recognition (PR) and remains a compelling topic in the study of computer vision. However, in real-world FR systems, interferences in images, including illumination condition, occlusion, facial expression and pose variation, make the recognition task challenging. This study explored the impact of those interferences on FR performance and attempted to alleviate it by taking face symmetry into account. A novel and robust FR method was proposed by combining multi-mirror symmetry with local binary pattern (LBP), namely multi-mirror local binary pattern (MMLBP). To enhance FR performance with various interferences, the MMLBP can 1) adaptively compensate lighting under heterogeneous lighting conditions, and 2) generate extracted image features that are much closer to those under well-controlled conditions (i.e., frontal facial images without expression). Therefore, in contrast with the later variations of LBP, the symmetrical singular value decomposition representation (SSVDR) algorithm utilizing the facial symmetry and a state-of-art non-LBP method, the MMLBP method is shown to successfully handle various image interferences that are common in FR applications without preprocessing operation and a large number of training images. The proposed method was validated with four public data sets. According to our analysis, the MMLBP method was demonstrated to achieve robust performance regardless of image interferences.

Face recognition has attracted great interest due to its importance in many real-world applications. In this paper, we present a novel low-rank sparse representation-based classification (LRSRC) method for robust face recognition. Given a set of test samples, LRSRC seeks the lowest-rank and sparsest representation matrix over all training samples. Since low-rank model can reveal the subspace structures of data while sparsity helps to recognize the data class, the obtained test sample representations are both representative and discriminative. Using the representation vector of a test sample, LRSRC classifies the test sample into the class which generates minimal reconstruction error. Experimental results on several face image databases show the effectiveness and robustness of LRSRC in face image recognition.

There usually exist diverse variations in face images taken under uncontrolled conditions. Most previous work on face recognition focuses on particular variations and usually assume the absence of others. Such work is called controlled face recognition. Instead of the ‘divide and conquer’ strategy adopted by controlled face recognition, this paper presents one of the first attempts directly aiming at uncontrolled face recognition. The solution is based on Individual Stable Neural Network (ISNN) proposed in this paper. ISNN can map a face image into the so-called Individual Stable Space (ISS), the feature space that only expresses personal characteristics, which is the only useful information for recognition. There are no restrictions for the face images fed into ISNN. Moreover, unlike many other robust face recognition methods, ISNN does not require any extra information (such as view angle) other than the personal identities during training. These advantages of ISNN make it a very practical approach for uncontrolled face recognition. In the experiments, ISNN is tested on two large face databases with vast variations and achieves the best performance compared with several popular face recognition techniques.

Robust discriminative nonnegative dictionary learning for occluded face recognition

Selective generation of Gabor features for fast face recognition on mobile devices

Face recognition is one of the biometric identification methods with the highest potential. The existing face recognition algorithms relying on the texture information of face images are affected greatly by the variation of expression, scale and illumination. Whereas the algorithms based on the shape topology weaken the influence of illumination to some extent, but the impact of expression, scale and illumination on face recognition is still unsolved. To this end, we propose a new method for face recognition by integrating texture information with shape information, called biview face recognition algorithm. The texture models are constructed by using subspace learning methods and shape topologies are formed by building graphs for face images. The proposed biview face recognition method is compared with recognition algorithms merely based on texture or shape information. Experimental results of recognizing faces under the variation of illumination, expression and scale demonstrate that the performance of the proposed biview face recognition outperforms texture-based and shape-based algorithms.

Pixel selection based on discriminant features with application to face recognition

Face recognition is efficiently used in computer vision, pattern recognition, biometric, law enforcement, surveillance, criminal investigations, and missing person detection. However, face recognition is still a challenging task under variant illuminations, pose, expression, occlusion, and background. In this paper, we proposed a cascading of histogram of oriented gradients (HOG) and local binary pattern (LBP) feature extraction method on ORL face dataset, which can improve recognition rates while also addressing pose, scale, expression, and variant illumination issues. The facial recognition network was trained using various classifiers such as with KNN, SVM, and random forest (RF). The results have been shown that the accuracy rate of the combined LBP and HOG feature extraction method is better than individual LBP or HOG features for ORL database face recognition.

Data augmentation has been utilized to improve the accuracy and robustness of face recognition algorithms. However, most of the previous studies focused on using the augmentation techniques to enlarge the feature set, while the diversity produced by the virtual samples lacked sufficient attention. In sparse dictionary learning-based face recognition, $$l_1$$ -based sparse representation (SR) and SVD-based dictionary learning (DL) both have shown promising performance. How to utilize both of them in an enhanced training process by data augmentation is still unclear. This paper proposes a novel method that utilizes the sample diversity generated by data augmentation and integrates sparse representation with dictionary learning, to learn dual sparse features for robust face recognition. An additional feature set is created by applying sample augmentation via simply horizontal flipping of face images. The two sparse models, $$l_1$$ -based SR and SVD-based DL, are integrated together using our new proposed objective function. Under two-level fusion of both data and classifiers, the diversity between two training sets is well learned and utilized, in three implementations, to obtain a robust face recognition. After conducting extensive experiments on some popular facial datasets, we demonstrate the proposed method can produce a higher classification accuracy than many state-of-the-art algorithms, and it can be considered as a promising option for image-based face recognition. Our code is released at GitHub.

In this paper, we propose a novel robust face recognition framework named nuclear norm based superposed collaborative representation classifier (NNSCRC) to handle illumination variations, occlusion and undersampled problems in face recognition. Specifically, we develop a superposed linear collaborative representation classifier for robust face recognition by representing the query image in terms of a superposition of the class centroid, the shared intra-class difference, and the low rank error. By representing a face image as the class centroid and the shared intra-class difference, our model can effectively enhance the face recognition performance on undersampled databases. In addition, since the occlusion and illumination variations generally lead to a low-rank error image, we use nuclear norm matrix regression to obtain these low-rank errors, which makes our model able to reconstruct the test image better. Extensive experiments are performed on Extended Yale-B and AR databases, which show the effectiveness of NNSCRC in robust face recognition.

At present, the performance of face recognition system depends much on the variations in illumination. To solve this problem, this paper presents an adaptable face recognition approach that uses filter fusion representation. The key idea is to use context-aware filter fusion to get better image from a bad illumination one. Genetic algorithm is the tool for adaptation for individual context category. These can provide robust face recognition on illumination context-awareness under uneven environments. Gabor wavelet representation can also provide a robust feature for image enhancement. Using these approaches, we have developed a robust face recognition technique that can recognize with a notable success and it has been tested on Inha DB and FERET face images.