A theorem on the uncorrelated optimal discriminant vectors

Yang, Jingyu; Tang, Zhenmin; Hu, Zhong-Shan

doi:10.1016/s0031-3203(00)00135-7

Cited by 103 publications

(42 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Four feature projection methods based on linear structure are evaluated in this study including ULDA [29], orthogonal LDA (OLDA) [30], OFNDA [17] and linear discriminant analysis via QRdecomposition (LDA/QR) [31].…”

Section: Discriminant Analysis Based Projection Methodsmentioning

confidence: 99%

Application of Linear Discriminant Analysis in Dimensionality Reduction for Hand Motion Classification

Phinyomark¹,

Hu²,

Phukpattaranont³

et al. 2012

Measurement Science Review

View full text Add to dashboard Cite

The classification of upper-limb movements based on surface electromyography (EMG) signals is an important issue in the control of assistive devices and rehabilitation systems. Increasing the number of EMG channels and features in order to increase the number of control commands can yield a high dimensional feature vector. To cope with the accuracy and computation problems associated with high dimensionality, it is commonplace to apply a processing step that transforms the data to a space of significantly lower dimensions with only a limited loss of useful information. Linear discriminant analysis (LDA) has been successfully applied as an EMG feature projection method. Recently, a number of extended LDA-based algorithms have been proposed, which are more competitive in terms of both classification accuracy and computational costs/times with classical LDA. This paper presents the findings of a comparative study of classical LDA and five extended LDA methods. From a quantitative comparison based on seven multi-feature sets, three extended LDA-based algorithms, consisting of uncorrelated LDA, orthogonal LDA and orthogonal fuzzy neighborhood discriminant analysis, produce better class separability when compared with a baseline system (without feature projection), principle component analysis (PCA), and classical LDA. Based on a 7-dimension time domain and time-scale feature vectors, these methods achieved respectively 95.2% and 93.2% classification accuracy by using a linear discriminant classifier.

show abstract

Section: Discriminant Analysis Based Projection Methodsmentioning

confidence: 99%

Application of Linear Discriminant Analysis in Dimensionality Reduction for Hand Motion Classification

Phinyomark¹,

Hu²,

Phukpattaranont³

et al. 2012

Measurement Science Review

View full text Add to dashboard Cite

show abstract

“…We then implement the SRC on the extracted-features yielded by using some popular dimension reduction approaches, i.e., PCA, LDA, Uncorrelated LDA (ULDA) [24], Orthogonal LDA (OLDA) [25], and Semidefinite Programming LDA (SDP) [26]. These methods can reduce feature dimension significantly in order to guarantee the efficiency of our system.…”

Section: Algorithm 1: Sparse Representation-based Classificationmentioning

confidence: 99%

Robust Face Recognition via Accurate Face Alignment and Sparse Representation

Wang

Shen

2010

2010 International Conference on Digital Image Computing: Techniques and Applications

View full text Add to dashboard Cite

“…Because of the conjugated orthogonality constraints (Equation (10)), any two of the optimal discriminant vectors set are statistically uncorrelated [13][14][15]. In Reference [14], when solving the (r + 1)th optimal discriminant vector w r+ 1 , on the one hand, it let w r+ 1 meet normalized condition.…”

Section: Suodv Algorithmmentioning

confidence: 99%

“…The Foley-Sammon optimal discriminant vectors are orthogonal to each other, but the discriminant features obtained by projecting the original features on these orthogonal optimal discriminant vectors are statistically correlated. Jin put forward the analytic algorithm of statistically uncorrelated optimal discriminant vectors (SUODV) [13,14], Wu presented an improved SUODV [15]. The improvement of these algorithms is that single optimal discriminant vector is changed into optimal vector set that meet a variety of constraint condition.…”

Section: Introductionmentioning

confidence: 99%

Partial discharge pattern recognition based on optimal uncorrelated discriminant vectors in GIS

Zhang

Yao

Tang

2008

Int Trans Elec Energy Syst

View full text Add to dashboard Cite

SUMMARYWhile partial discharge (PD) pattern recognition is widely considered as an important method to evaluate insulation state, the major problem encountered in recognition is to extract PD pattern features. This paper designed four typical physical models of defects in GIS, and PD threedimensional patterns were constructed based on mass sample gathered by the very high frequency (VHF) and high-speeds systems. A principal component analysis (PCA)-statistically uncorrelated optimal discriminant vectors (SUODV) method is put forward based on 3D patterns images. Firstly, the PCA is employed to condense the dimension of PD images, then uncorrelated discriminate features are extracted by improved SUODV algorithm, and the minimum distance classifier was constructed as classifier. The recognition results showed that this method can effectively classify four kinds of defects in GIS.

show abstract

A theorem on the uncorrelated optimal discriminant vectors

Cited by 103 publications

References 16 publications

Application of Linear Discriminant Analysis in Dimensionality Reduction for Hand Motion Classification

Application of Linear Discriminant Analysis in Dimensionality Reduction for Hand Motion Classification

Robust Face Recognition via Accurate Face Alignment and Sparse Representation

Partial discharge pattern recognition based on optimal uncorrelated discriminant vectors in GIS

Contact Info

Product

Resources

About