Face recognition algorithm based on discriminative dictionary learning and sparse representation

An automatic segmentation method of hippo-campus for volume measurement in MR brain images by using sparse patch representation and discriminative dictionary learning is proposed in this paper, which can overcome the limitation of multi-atlas approaches that mostly rely on similarity between target image and atlases for more accurate segmentation. In the proposed method, atlases are registered to a target image, then patch-based features including intensity and gradient are extracted from registered atlases for a dictionary learning model, from which dictionaries and classifiers are learned simultaneously. The label for voxels in the target image is determined via solving sparse representation of target patch over learned dictionary. Performance of the proposed method is evaluated on 30 subjects from the public dataset CIND by cross validation. Qualitative and quantitative comparisons illustrate that the proposed method is competitive compared with state-of-the-art approaches based on multi-atlas.

Joint discriminative and shared dictionary learning with dictionary extension strategy for bearing fault classification

The sparse representation based classification (SRC) can effectively improve the face recognition rate. Smoothed l 0 algorithm has much faster calculation speed and requires fewer measured values than the other sparse representation method. In this paper, the sparse representation and smoothed l 0 algorithm for face recognition are presented to improve the face recognition under various conditions such as face disguise, illumination and pose changes, etc. The experiments on the AR, Extended Yale B and FERET face database verify the effectiveness of the presented method. The experimental results show that the face recognition algorithm increases to a certain extent in terms of recognition robustness and time than the original SRC.

Both interclass variances and intraclass similarities are crucial for improving the classification performance of discriminative dictionary learning (DDL) algorithms. However, existing DDL methods often ignore the combination between the interclass and intraclass properties of dictionary atoms and coding coefficients. To address this problem, in this paper, we propose a discriminative Fisher embedding dictionary learning (DFEDL) algorithm that simultaneously establishes Fisher embedding models on learned atoms and coefficients. Specifically, we first construct a discriminative Fisher atom embedding model by exploring the Fisher criterion of the atoms, which encourages the atoms of the same class to reconstruct the corresponding training samples as much as possible. At the same time, a discriminative Fisher coefficient embedding model is formulated by imposing the Fisher criterion on the profiles (row vectors of the coding coefficient matrix) and coding coefficients, which forces the coding coefficient matrix to become a block-diagonal matrix. Since the profiles can indicate which training samples are represented by the corresponding atoms, the proposed two discriminative Fisher embedding models can alternatively and interactively promote the discriminative capabilities of the learned dictionary and coding coefficients. The extensive experimental results demonstrate that the proposed DFEDL algorithm achieves superior performance in comparison with some state-of-the-art dictionary learning algorithms on both hand-crafted and deep learning-based features.

For face recognition, conventional dictionary learning (DL) methods have disadvantages. In the paper, we propose a novel robust, discriminative and comprehensive DL (RDCDL) model. The proposed model uses sample diversities of the same face image to make the dictionary robust. The model includes class-specific dictionary atoms and disturbance dictionary atoms, which can well represent the data from different classes. Both the dictionary and the representation coefficients of data on the dictionary introduce discriminative information, which improves effectively the discrimination capability of the dictionary. The proposed RDCDL is extensively evaluated on benchmark face image databases, and it shows superior performance to many state-of-the-art sparse representation and dictionary learning methods for face recognition.

Although the object detection and recognition has received growing attention for decades, a robust fire and flame detection method is rarely explored. This paper presents a novel solid solution based on sparse representation with spatial pyramid pooling. Traditional sparse representation, as one of prevalent feature learning methods, is successfully applied for object detection. But it has some intrinsic defects. Firstly, it requires fixed input image size. Secondly, the accuracy of detection heavily depends on discriminative dictionary learning and feature coding. At last, it is usually very time-consuming. In this paper, we have proposed a novel dictionary learning method with the structure sparsity constraint to train a discriminative dictionary. In feature coding stage, we compute sparse codes of each patch with dictionaries learned from data and pool them to form local histogram in spatial pyramid manner. At lat, the feature vector is pipelined into a linear SVM classifier to train the model. For improving the efficiency, we also adopt the selective search approach to generate the candidate region proposals in the preprocessing stage. In processing test images, our method achieved better or comparable accuracy to the state-of-the-art on FlameDetection2010 Dataset.

Robust discriminative nonnegative dictionary learning for occluded face recognition

Face recognition is a challenging issue in field of multi-science, main contents of research is how to make computer have ability of face recognition face recognition technology involved in a lot, which is a key feature extraction and classification method, this paper focuses on study of related theory. Non-negative matrix factorization trapped MF) algorithm and local non-negative matrix factorization (LNMF) algorithm is a feature extraction method based on local features, has been successfully used in face recognition NMF algorithm in face recognition rate is low, although LNMF algorithm to a certain extent, improve recognition rate, but its price is to increase number of iterations. In addition, two algorithms have failed to solve good nonlinear separable problems the kernel method combined with LNMF algorithm, kernel local non-negative matrix factorization (KLNMF) algorithm, first by a nonlinear transformation of original space to high-dimensional space, making samples linearly separable, and then use LNMF algorithm to extract face features. In classification part, paper presents decision rules of classification of their own, and design based on NMF subspace classifier. DOI : http://dx.doi.org/10.11591/telkomnika.v12i3.4356 Full Text: PDF

In this thesis we present two applications where 3D surface modeling and registration play a role of paramount importance. The first application is in archaeology, where our proposed method assists the archaeologist in the tedious process of reconstructing ceramic vessels from the excavated fragments (or shards). The method takes advantage of the surface intrinsic features coupled with a library of generic models constructed by the archaeologists to produce a virtual reconstruction of what the original vessel may have looked like. Generic models are generated based on the expert historical knowledge of the period, provenance of the artifact, and location of the digging site. The generic models need not to be identical to the original vessel, but must be within a geometric transformation of it in most of its parts. The methods are suited for ceramic vessels with distinctive color patterns or reliefs (i.e., surface with molding, carving, or stamping). We introduce a novel curvature B-spline fitting method that makes the extraction of robust features possible. The alignment of the shards against the generic model uses a novel set of 3D weighted curve moments and absolute affine invariants derived from them. The transformation is computed from corresponding parabolic contours on the shard and the generic model. When the shards have no color or relief on them, they are virtually mended to abutting shards using intrinsic differential anchor points computed on the shards break curves and a set of absolute invariants. For axially symmetric vessels, a global constraint induced by the surface of revolution is applied to guarantee global mending consistency. In our experiment involving around 300 shards, more than 85% of them are successfully mended. The mended shards consist of over 90% surface areas of the ceramic vessels. The second application is in 3D human face synthesis and recognition, where our proposed method assists the national security agencies in constructing a 3D face database to be used for spotting missing personnel or identify possible suspects when close up images of that person are flagged up. A gender/race generic 3D face model is morphed into a personalized unique 3D face of a given person from using a limited number of 2D images of the given person taken from different viewpoints. The 2D images need not to be of the same resolution, or taken under similar brightness or same place, or even same time period. The method applies also to drawing sketches of the person in question. The reconstructed 3D personalized model is an accurate rendition of the actual human face, and the reconstruction errors are smaller than a 3D scanner resolution that is used to scan the person's face. Our method renders a difficult 2D face recognition problem into a much simpler 3D surface registration (alignment) problem. Given a set of images of a test person, a personalized 3D face model is constructed and "best" aligned with one of the 3D faces stored in the database. When the minimum alignment errors are below an accepted threshold (lower than the scanner resolution), the identity of the test person is matched with the one in the database. Experimental results are shown in controlled environment with high resolution images, under ideal illumination condition, as well as for images taken in uncontrolled environment with low resolution, arbitrary illumination. In both conditions, the recognition rate is over 95%. It is also proved that better face recognition rate is achieved by using a gender/race based generic 3D face model which allows for accurate pose estimation as well as the face identification in 3D.

This paper represents a new approach for face recognition that incorporates Prewitt edge detection, Gabor filter and Zernike moments to transform the image into a unified domain. On this joint domain, five distance metrics are constructed using Schoenberg transform for the purpose of defining efficient similarity measures for holistic face recognition. The proposed Schoenberg similarity applies Schoenberg transform to the logarithm of five existing distance metrics: Minkowski, City-Block, Euclidean, Soergel and Lorentzian metrics. The constructed Schoenberg logarithmic metrics are called SL-Minkowski, SL-City-Block, SL-Euclidean, SL-Soergel and SL-Lorentzian. These distance metrics are utilized as similarity measures after being normalized over the range [0,1] for fair comparison with existing measures. The proposed Schoenberg system can resist three problems: Change in illumination, pose and facial expression. Simulation results show that the proposed distance measures have superior performance as compared to the classical metrics: Structural Similarity Index Measure (SSIM) and Feature-based Similarity Measure (FSIM). Performance criteria are the recognition rate and the recognition confidence, defined as the similarity difference between the best match and the second-best match in the face database.

We address the problem of sparse channel representation for downlink channel estimation in multi-user frequency division duplexing (FDD) massive multiple-input multiple-output (MIMO) systems. Existing methods typically adopt discrete Fourier transform (DFT) matrix as a sparse basis to represent sparse channels. However, a sparse basis constructed through dictionary learning method has proven to have strong sparse channel representation. In this work, we develop a discriminative dictionary learning-based sparse channel representation for downlink channel estimation in a multi-user FDD massive MIMO systems. Considering partially shared support between near users, we present a new discriminative dictionary learning (DDL) method for sparse channel representation, based on which the channel estimation scheme is developed. Compared with learning a shared dictionary for all users, it can provide a better representation, thus improving the performance of recovery in the compressive sensing process. Numerical results demonstrate the superior performance of discriminative dictionary as compared to the shared dictionary in terms of normalized mean square error (NMSE) and symbol error rate (SER).

It is known that face recognition rate is affected by the resolution of probe image. Therefore it is natural to expect improving the resolution via face hallucination would increase the recognition rate. However, it was concluded in the previous works that improvement in the visual quality does not necessarily lead to a better recognition rate, and performance of face hallucination for recognition is especially poor when very low resolution (VLR) images are used. Experiment results in this paper will show that there is a correlation between visual quality and recognition rate. And hallucination space relationship learning algorithm is proposed which is robust on hallucinating VLR images to improve the performance in terms of both visual quality and recognition rate.

Current automatic facial recognition systems are not robust against changes in illumination, pose, facial expression and occlusion. In this paper, we propose an algorithm based on a probabilistic approach for face recognition to address the problem of pose change by a probabilistic approach that takes into account the pose difference between probe and gallery images. By using a large facial image database called CMU PIE database, which contains images of the same set of people taken from many different angles, we have developed a probabilistic model of how facial features change as the pose changes. This model enables us to make our face recognition system more robust to the change of poses in the probe image. The experimental results show that this approach achieves a better recognition rate than conventional face recognition methods over a much larger range of pose. For example, when the gallery contains only images of a frontal face and the probe image varies its pose orientation, the recognition rate remains within a less than 10% difference until the probe pose begins to differ more than 45 degrees, whereas the recognition rate of a PCA-based method begins to drop at a difference as small as 10 degrees, and a representative commercial system at 30 degrees.

An efficient face recognition system should recognize faces in different views and poses. The efficiency of a human face recognition system depends on the capability of face recognition in presence of changes in the appearance of face due to expression, pose and illumination. A novel algorithm which utilizes the combination of texture and depth information based on Modular PCA to overcome the problem of pose variation and illumination change for face recognition is proposed. The system has combined 2D and 3D systems in the feature level which presents higher performance in contrast with methods which utilizes either 2D or 3D system separately. A multimodal face recognition based on Modular PCA when compared with conventional PCA algorithm has an improved recognition rate for face images with large variations in illumination and facial expression. The proposed algorithm is tested with FRAV3D database that has faces with pose variation and illumination changes. Recognition rates from experimental results show the superiority of Modular PCA over conventional PCA methods in tackling face images with different pose variations and changes in illuminations. The proposed algorithm shows a recognition rate of 86% that is achieved in fusion experiment.

Discriminative dictionary learning (DDL) framework has been widely used in image classification which aims to learn some class-specific feature vectors as well as a representative dictionary according to a set of labeled training samples. However, interclass similarities and intraclass variances among input samples and learned features will generally weaken the representability of dictionary and the discrimination of feature vectors so as to degrade the classification performance. Therefore, how to explicitly represent them becomes an important issue. In this paper, we present a novel DDL framework with two-level low rank and group sparse decomposition model. In the first level, we learn a class-shared and several class-specific dictionaries, where a low rank and a group sparse regularization are, respectively, imposed on the corresponding feature matrices. In the second level, the class-specific feature matrix will be further decomposed into a low rank and a sparse matrix so that intraclass variances can be separated to concentrate the corresponding feature vectors. Extensive experimental results demonstrate the effectiveness of our model. Compared with the other state-of-the-arts on several popular image databases, our model can achieve a competitive or better performance in terms of the classification accuracy.