Kernel Sparse Representation-Based Classifier

Zhang, Li; Zhou, Wenjing; Chang, Pei‐Chann; Jing, Liu; Yan, Zhe; Wang, Ting; Li, Fanzhang

doi:10.1109/tsp.2011.2179539

Cited by 300 publications

(155 citation statements)

References 34 publications

Supporting

Mentioning

154

Contrasting

Unclassified

Order By: Relevance

“…The sample membership functions can be got by the rule of fuzzy neighbour [10]: First of all we make up the distance matrix 1 D by the distance between the training samples. Second, we set the second diagonal matrix 1 D to infinity; then by size order matrix of each line, we can get the matrix 2 D , by which K nearest neighbor points and the corresponding category information can be obtained. Finally by (1), the corresponding membership degree matrix P is the priori probability of i th class sample, which is defined as follows:…”

Section: A Two-dimensional Fuzzy Basic Idea Of Kernel Principal Compmentioning

confidence: 99%

“…Face recognition has become an important research direction of biometrics, face feature description is one of the key steps [1], mainly by extracting the effective face identification and the design of complex classifier [2]. Principal component analysis (PCA) [3] and two dimensional principal component analysis of (2DPCA) [4] are both the linear feature extraction method of minimum square error sense.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fuzzy Kernel Two-dimensional Principal Component Analysis for Face Recognition

Zeng¹,

Chen²,

Tian³

et al. 2015

Advances in Intelligent Systems Research

View full text Add to dashboard Cite

-The traditional kernel two-dimensional principal component analysis (K2DPCA) method did not take full advantage of the class information for face images and there are both "outer class" problem and "hard classifier" problem on face recognition. Therefore, a new face recognition method based on fuzzy kernel two-dimensional principal component analysis (FK2DPCA) is presented . Firstly, it introduces fuzzy concept into K2DPCA. Secondly, the class separability of criterion will be extended to high dimensional feature space by the use of kernel method. Furthermore, we select the eigenvectors that betweenclass scatter is greater than within-class scatter after projection as optimal projection axis. Finally, it uses the nearest neighbor classifier for face recognition . The experiment results on ORL and YALE face databases show that the FK2DPCA is better than other traditional methods.

show abstract

Section: A Two-dimensional Fuzzy Basic Idea Of Kernel Principal Compmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Fuzzy Kernel Two-dimensional Principal Component Analysis for Face Recognition

Zeng¹,

Chen²,

Tian³

et al. 2015

Advances in Intelligent Systems Research

View full text Add to dashboard Cite

show abstract

“…According to [1] the classifier shows the good performance results on face image data in which they combined the result of machine leaning and compressed sensing. This paper overcomes the drawback of sparse representation classifier.…”

Section: Related Workmentioning

confidence: 99%

Kernel Collaboration with Perceptron for Facial Emotion Recognition

Ramachandran¹,

Thaiyalnayaki²

2013

IJCA

View full text Add to dashboard Cite

Many classifiers are in existence for handling linear and nonlinear type of data.In this paper, a method for handling nonlinear data for classification is proposed. The classifier is modeled for both two-class and multi-class problem. In this framework, the input samples are projected into a higher dimensional feature space by utilizing kernel trick. In order to avoid the difficulty of working in this higher dimension space, a dimensional reduction technique is used. This reduction technique results with the transformation matrix which is of lesser dimensions. Then the classification of data is carried out in this reduced subspace by using perceptron technique. The classifier is utilized for recognizing the emotion of a person based on the facial expressions. Experiments are carried on toy data set and JAFFE data set and also the results have shownthe effectiveness of this classifier. The classifier recognizes the 6 different Emotions with 98.6% efficiency.

show abstract

“…Several other methods have also been proposed that essentially exploit the non-linear structure of data by sparse coding in the feature space [12], [13]. In [14], Yuan and Yan propose a multi-task joint sparse representation for visual recognition.…”

Section: Introductionmentioning

confidence: 99%

“…Their method is formulated as the solution to the problem of multitask least squares regression problem with ℓ 1,2 mixed-norm regularization. In [13], Zhang et al propose a kernel version of the sparse representation-based classification algorithm which was originally proposed for robust face recognition [15]. The optimization approach presented in [11] is purely generative.…”

Section: Introductionmentioning

confidence: 99%

Design of Non-Linear Discriminative Dictionaries for Image Classification

Shrivastava

Nguyen

Patel

et al. 2013

Computer Vision – ACCV 2012

View full text Add to dashboard Cite

Abstract. In recent years there has been growing interest in designing dictionaries for image classification. These methods, however, neglect the fact that data of interest often has non-linear structure. Motivated by the fact that this non-linearity can be handled by the kernel trick, we propose learning of dictionaries in the high-dimensional feature space which are simultaneously reconstructive and discriminative. The proposed optimization approach consists of two main stages-coefficient update and dictionary update. We propose a kernel driven simultaneous orthogonal matching pursuit algorithm for the task of sparse coding in the feature space. The dictionary update step is performed using an efficient approximate KSVD algorithm in feature space. Extensive experiments on image classification demonstrate that the proposed non-linear dictionary learning method is robust and can perform significantly better than many competitive discriminative dictionary learning algorithms.

show abstract

Kernel Sparse Representation-Based Classifier

Cited by 300 publications

References 34 publications

Fuzzy Kernel Two-dimensional Principal Component Analysis for Face Recognition

Fuzzy Kernel Two-dimensional Principal Component Analysis for Face Recognition

Kernel Collaboration with Perceptron for Facial Emotion Recognition

Design of Non-Linear Discriminative Dictionaries for Image Classification

Contact Info

Product

Resources

About