Diagonal Fisher linear discriminant analysis for efficient face recognition

Noushath, S.; Kumar, G. Hemantha; Palaiahnakote, Shivakumara

doi:10.1016/j.neucom.2006.01.012

Cited by 42 publications

(13 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…By doing this, some considerable merits were achieved: i) Computation of scatter matrices became simple ii)Computation time also reduced and iii) significant improvement in accuracy was achieved as compared with their 1-D counterparts [12]. Recently, method namely DiaFLD [13] have been proposed. This method seek optimal projection vectors from diagonal character image so that the correlation between variations of both rows and columns of images can be preserved.…”

Section: B Some Extended Version Of Fld Methodsmentioning

confidence: 99%

A Robust Analysis of FLD and Orthogonal FLD on Handwritten Characters

Aradhya¹,

Niranjan²,

Hamsaveni

2013

2013 International Conference on Communication Systems and Network Technologies

View full text Add to dashboard Cite

Feature extraction is the identification of appropriate measures to characterize the component images distinctly.Extracting features is one of the most important steps in any recognition system. Hence, in this paper, we explore the concept of Orthogonalized Fisher Discriminant (OFD) for unconstrained handwritten Kannada character recognition. OFD exhibits higher performance than Fisher Linear Discriminant (FLD) due to the elimination of dependences among discriminant vectors. For subsequent classification purpose, we explore the concept of probabilistic neural network (PNN) architecture. Experiments show that OFD methods are more effective and efficient than standard FLD for handwritten character recognition.

show abstract

Section: B Some Extended Version Of Fld Methodsmentioning

confidence: 99%

A Robust Analysis of FLD and Orthogonal FLD on Handwritten Characters

Aradhya¹,

Niranjan²,

Hamsaveni

2013

2013 International Conference on Communication Systems and Network Technologies

View full text Add to dashboard Cite

show abstract

“…Both DQDA and DLDA are easy to implement and have been widely adopted to analyze high-dimensional data in the fields of science. See for example Speed (2003), Ye et al (2004), Noushath et al (2006), Asyali et al (2006), Son and Lee (2006), Heilemann and Schuhr (2008), Natowicz et al (2008), among many others.…”

Section: Diagonal Discriminant Analysis and Its Improvementmentioning

confidence: 99%

Block-diagonal discriminant analysis and its bias-corrected rules

Pang

Tong

2013

Statistical Applications in Genetics and Molecular Biology

View full text Add to dashboard Cite

High-throughput expression profiling allows simultaneous measure of tens of thousands of genes at once. These data have motivated the development of reliable biomarkers for disease subtypes identification and diagnosis. Many methods have been developed in the literature for analyzing these data, such as diagonal discriminant analysis, support vector machines, and k-nearest neighbor methods. The diagonal discriminant methods have been shown to perform well for high-dimensional data with small sample sizes. Despite its popularity, the independence assumption is unlikely to be true in practice. Recently, a gene module based linear discriminant analysis strategy has been proposed by utilizing the correlation among genes in discriminant analysis. However, the approach can be underpowered when the samples of the two classes are unbalanced. In this paper, we propose to correct the biases in the discriminant scores of block-diagonal discriminant analysis. In simulation studies, our proposed method outperforms other approaches in various settings. We also illustrate our proposed discriminant analysis method for analyzing microarray data studies.

show abstract

“…Noushath 등은 Dia2DPCA에서 아이디어를 얻어서 대각 방향 얼굴을 이용한 Dia2DLDA (Diagonal two-dimensional LDA)를 제안했다 [13]. 또한, 특징 행렬의 크기를 줄 이기 위해서 Dia2DLDA에 2DLDA를 결합하였다.…”

Section: 대각 2dlda(dia2dlda)unclassified

Bilateral Diagonal 2DLDA Method for Human Face Recognition

Kim¹,

Song²,

Kim³

et al. 2009

Journal of Korean Institute of Intelligent Systems

View full text Add to dashboard Cite

In this paper, a method called bilateral diagonal 2DLDA is proposed for face recognition. Two methods called Dia2DPCA and Dia2DLDA were suggested to reserve the correlations between the variations in the rows and columns of diagonal images. However, these methods work in the row direction of these images. A row-directional projection matrix can be obtained by calculating the between-class and within-class covariance matrices making an allowance for the column variation of alternative diagonal face images. In addition, column-directional projection matrix can be obtained by calculating the between-class and within-class covariance matrices making an allowance for the row variation in diagonal images. A bilateral projection scheme was applied using left and right multiplying projection matrices. As a result, the dimension of the feature matrix and computation time can be reduced. Experiments carried out on an ORL face database show that the proposed method with three different distance measures, namely, Frobenius, Yang and AMD, is more accurate than some methods, such as 2DPCA, B2DPCA, 2DLDA, etc.

show abstract

Diagonal Fisher linear discriminant analysis for efficient face recognition

Cited by 42 publications

References 6 publications

A Robust Analysis of FLD and Orthogonal FLD on Handwritten Characters

A Robust Analysis of FLD and Orthogonal FLD on Handwritten Characters

Block-diagonal discriminant analysis and its bias-corrected rules

Bilateral Diagonal 2DLDA Method for Human Face Recognition

Contact Info

Product

Resources

About