Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) are well known techniques for face recognition. Both PCA and LDA by themselves have good recognition rates. We propose Canonical Correlation Analysis (CCA) for combining two feature extractors to improve the performance of the system, by obtaining the advantages of both. CCA finds the transformation for each extractor dat...

We present a fast algorithm for approximate canonical correlation analysis (CCA). Given a pair of tall-and-thin matrices, the proposed algorithm first employs a randomized dimensionality reduction transform to reduce the size of the input matrices, and then applies any CCA algorithm to the new pair of matrices. The algorithm computes an approximate CCA to the original pair of matrices with prov...

Partial least squares (PLS) approach is proposed for linear discriminant analysis (LDA) when predictors are data of functional type (curves). Based on the equivalence between LDA and the multiple linear regression (binary response) and LDA and the canonical correlation analysis (more than two groups), the PLS regression on functional data is used to estimate the discriminant coefficient functio...

We review a recent neural implementation of Canonical Correlation Analysis and show, using ideas suggested by Ridge Regression, how to make the algorithm robust. The network is shown to operate on data sets which exhibit multicollinearity. We develop a second model which not only performs as well on multicollinear data but also on general data sets. This model allows us to vary a single paramet...

154 specimens of periophthalmus waltoni distributed in seven different sites in the persian gulf and gulf of oman are analysed, utilizing ten morphometric characters. canonical variate analysis (cva) of the sampling sites suggested a partial separation between the north-western sites and the central plus south-eastern sites. the main factor to connect p. waltoni specimens from different sites a...

In this article an accurate and efficient technique for tissue typing is presented. The proposed technique is based on Canonical Correlation Analysis, a statistical method able to simultaneously exploit the spectral and spatial information characterizing the Magnetic Resonance Spectroscopic Imaging (MRSI) data. Recently, Canonical Correlation Analysis has been successfully applied to other type...

in order to study the variation of fatty acids and protein content of soybean seeds and correlated morphological and phenological traits by canonical correlation analysis, 25 genotypes were evaluated with latis square design in baieakola agricultural research station of mazandaran agricultural research center in 2001. a number of 20 agronomic and eight quality traits were measured. the variatio...