(2D)2 DLDA for Efficient Face Recognition

Cho, Dong-Uk; Chang, Un-Dong; Kim, Kwan-Dong; Kim, Bong-hyun; Lee, Se-Hwan

doi:10.1007/11949534_31

Cited by 4 publications

(3 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Specially, the blocks are the row directional lines of the images. Inspired by 2DPCA, some other methods based on 2D image matrix were proposed, such as 2DLDA, 2D-DLDA, (2D)2DLDA and two dimensional Laplacianfaces method [7]- [11].…”

Section: Introductionmentioning

confidence: 99%

A Simulated Annealing and 2DPCA Based Method for Face Recognition

Wang

Yang

2007

2007 Third International IEEE Conference on Signal-Image Technologies and Internet-Based System

View full text Add to dashboard Cite

In this paper we address the problem of face recognition based on two-dimensional principal component analysis (2DPCA). The similarity measure plays an important role in pattern recognition. However, with reference to the 2DPCA based method for face recognition, studies on similarity measures are quite few. We propose a new method to identify the similarity measure by simulated annealing (SA), which is called SA similarity measure. Experimental results on two famous face databases show that the proposed method outperforms the state of the art methods in terms of recognition accuracy.

show abstract

Section: Introductionmentioning

confidence: 99%

A Simulated Annealing and 2DPCA Based Method for Face Recognition

Wang

Yang

2007

2007 Third International IEEE Conference on Signal-Image Technologies and Internet-Based System

View full text Add to dashboard Cite

show abstract

“…Since LDA often suffers from the small sample size (3S) problem, some effective approaches have been proposed, such as PCA + LDA [1], orthogonal LDA [42], LDA/GSVD [11], and LDA/QR [43]. Recently, 2DLDA and its variants have attracted much attention from researchers due to the advantages over the singularity problem and the computational cost (e.g., [8], [18]- [20], [35], [39]). As we know that LDA-like methods is to seek the discriminant vectors such that the ratio of the between-class distance to the within-class distance is maximized with applying the label information.…”

Section: Introductionmentioning

confidence: 99%

Improved Two-Dimensional Quaternion Principal Component Analysis

2019

View full text Add to dashboard Cite

The two-dimensional quaternion principal component analysis (2D-QPCA) is first improved into abstracting the features of quaternion matrix samples in both row and column directions, being the generalization ability, and with the components weighted by corresponding eigenvalues. The main components of the 2D-QPCA in row and column directions are defined as the solution of one optimal problem with orthogonality constraints on two variables. In advance, 2D-QPCA is innovatively armed with the generalization ability by applying the label information (if known) to generate the weight of each class of training set. With such generalization ability, 2D-QPCA may enlarge the variance of all projections of known and unknown samples. Different from the well-known PCA-like methods, the improved 2D-QPCA prefers to the components that have larger variances of projected samples and weight them with larger factors. The improved versions of 2D-QPCA are applied to the color face recognition and image reconstruction. The numerical results based on the real face data sets validate that the newly proposed method performs better than the state-of-the-art ones.

show abstract

“…Since LDA often suffers from the small sample size (3S) problem, some effective approaches have been proposed, such as PCA + LDA [35], orthogonal LDA [36], LDA/GSVD [37], and LDA/QR [38]. Because of the advantages over the singularity problem and the computational cost, 2DLDA and its variants have recently attracted much attention from researchers (e.g., [39,40,41,42,43,44]). With applying the label information, the LDA-like methods are intend to compute the discriminant vectors which maximize the ratio of the between-class distance to the within-class distance.…”

Section: Introductionmentioning

confidence: 99%

Advanced Variations of Two-Dimensional Principal Component Analysis for Face Recognition

Zhao

Jia

Cai

et al. 2019

Preprint

View full text Add to dashboard Cite

The two-dimensional principal component analysis (2DPCA) has become one of the most powerful tools of artificial intelligent algorithms. In this paper, we review 2DPCA and its variations, and propose a general ridge regression model to extract features from both row and column directions. To enhance the generalization ability of extracted features, a novel relaxed 2DPCA (R2DPCA) is proposed with a new ridge regression model. R2DPCA generates a weighting vector with utilizing the label information, and maximizes a relaxed criterion with applying an optimal algorithm to get the essential features. The R2DPCA-based approaches for face recognition and image reconstruction are also proposed and the selected principle components are weighted to enhance the role of main components. Numerical experiments on well-known standard databases indicate that R2DPCA has high generalization ability and can achieve a higher recognition rate than the state-of-the-art methods, including in the deep learning methods such as CNNs, DBNs, and DNNs.

show abstract

(2D)2 DLDA for Efficient Face Recognition

Cited by 4 publications

References 8 publications

A Simulated Annealing and 2DPCA Based Method for Face Recognition

A Simulated Annealing and 2DPCA Based Method for Face Recognition

Improved Two-Dimensional Quaternion Principal Component Analysis

Advanced Variations of Two-Dimensional Principal Component Analysis for Face Recognition

Contact Info

Product

Resources

About