A generalized optimal set of discriminant vectors

Liu, Ke; Cheng, Yujie; Yang, Jingyu

doi:10.1016/0031-3203(92)90136-7

Cited by 87 publications

(64 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…S W Null Space Method It has been introduced by Chen et al in [6] and exploits an important result of Liu et al [15] who showed that Fisher's criterion (14) is equivalent to:…”

Section: The Small Sample Size Problemmentioning

confidence: 99%

Enhancing Dimensionality Reduction Methods for Side-Channel Attacks

Cagli¹,

Dumas²,

Prouff³

2016

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Advanced Side-Channel Analyses make use of dimensionality reduction techniques to reduce both the memory and timing complexity of the attacks. The most popular methods to effectuate such a reduction are the Principal Component Analysis (PCA) and the Linear Discriminant Analysis (LDA). They indeed lead to remarkable efficiency gains but their use in side-channel context also raised some issues. The PCA provides a set of vectors (the principal components) onto which project the data. The open question is which of these principal components are the most suitable for side-channel attacks. The LDA has been valorized for its theoretical leaning toward the class-distinguishability, but discouraged for its constraining greed of data. In this paper we present an in-depth study of these two methods, and, to automatize and to ameliorate the principal components selection, we propose a new technique named cumulative Explained Local Variance (ELV) selection. Moreover we present some extensions of the LDA, available in less constrained situations than the classical version. We equip our study with a comprehensive comparison of the existing and new methods in real cases. It allows us to verify the soundness of the ELV selection, and the effectiveness of the methods proposed to extend the use of the LDA to side-channel contexts where the existing approaches are inapplicable.

show abstract

“…S W Null Space Method It has been introduced by Chen et al in [6] and exploits an important result of Liu et al [15] who showed that Fisher's criterion (14) is equivalent to:…”

Section: The Small Sample Size Problemmentioning

confidence: 99%

Enhancing Dimensionality Reduction Methods for Side-Channel Attacks

Cagli¹,

Dumas²,

Prouff³

2016

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Most LDAbased algorithms including Fisherfaces [20] and D-LDA [21] utilize the conventional Fisher criterion defined in (4) while some authors use the alternative given in (6) proposed by Liu [22,23].…”

Section: D-ldamentioning

confidence: 99%

Face recognition using nonparametric-weighted Fisherfaces

Prasad

Hsu

et al. 2012

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

This study presents an appearance-based face recognition scheme called the nonparametric-weighted Fisherfaces (NW-Fisherfaces). Pixels in a facial image are considered as coordinates in a high-dimensional space and are transformed into a face subspace for analysis by using nonparametric-weighted feature extraction (NWFE). According to previous studies of hyperspectral image classification, NWFE is a powerful tool for extracting hyperspectral image features. The Fisherfaces method maximizes the ratio of between-class scatter to that of within-class scatter. In this study, the proposed NW-Fisherfaces weighted the between-class scatter to emphasize the boundary structure of the transformed face subspace and, therefore, enhances the separability for different persons' face. The proposed NW-Fisherfaces was compared with Orthogonal Laplacianfaces, Eigenfaces, Fisherfaces, direct linear discriminant analysis, and null space linear discriminant analysis methods for tests on five facial databases. Experimental results showed that the proposed approach outperforms other feature extraction methods for most databases.

show abstract

“…Because of its emphasis on the discriminatory information content, the FoleySammon transform (FST) has been used extensively as one of the methods of dimensionality reduction in pattern recognition [9]- [12]. Various methods for determining FST under different conditions have been developed for human facial image recognition [13]- [16]. Liu proposed a modified class separability criterion, leading to a generalized optimal set of discriminant vectors [13].…”

Section: Introductionmentioning

confidence: 99%

“…Various methods for determining FST under different conditions have been developed for human facial image recognition [13]- [16]. Liu proposed a modified class separability criterion, leading to a generalized optimal set of discriminant vectors [13]. Several algorithms for determining the generalized optimal set of discriminant vectors were developed [14]- [16].…”

Section: Introductionmentioning

confidence: 99%

On Dimensionality Reduction for Client Specific Discriminant Analysis with Application to Face Verification

Kittler

Yang

et al. 2004

Advances in Biometric Person Authentication

View full text Add to dashboard Cite

Abstract. In this paper we propose a study on dimensionality reduction for client specific discriminant analysis with application to face verification. A new algorithm of face verification based on client specific discriminant analysis is developed. Two aspects of improvement are made in the new algorithm. First, a dimensionality reduction based on the between-class scatter matrix is introduced which is more efficient than that based on the population scatter matrix. The second improvement lies in the use of a new Fisher criterion function which is introduced in order to reduce the computational complexity of the client specific discriminant analysis problem. The experimental results obtained on the internationally recognized facial database XM2VTS using the Lausanne protocol show the effectiveness of the proposed method.

show abstract

A generalized optimal set of discriminant vectors

Cited by 87 publications

References 10 publications

Enhancing Dimensionality Reduction Methods for Side-Channel Attacks

Enhancing Dimensionality Reduction Methods for Side-Channel Attacks

Face recognition using nonparametric-weighted Fisherfaces

On Dimensionality Reduction for Client Specific Discriminant Analysis with Application to Face Verification

Contact Info

Product

Resources

About