Discriminative Feature Extraction by a Neural Implementation of Canonical Correlation Analysis.

Canonical correlation analysis using within-class coupling

Hotelling’s Canonical Correlation Analysis (CCA) works with two sets of related variables, also called views, and its goal is to find their linear projections with maximal mutual correlation. CCA is most suitable for unsupervised feature extraction when given two views but it has been also long known that in supervised learning when there is only a single view of data given, the supervision signal (class-labels) can be given to CCA as the second view and CCA simply reduces to Fisher’s Linear Discriminant Analysis (LDA). However, it is unclear how to use this equivalence for extracting features from multiview data in semisupervised setting (i.e. what modification to the CCA mechanism could incorporate the class-labels along with the two views of the data when labels of some samples are unknown). In this paper, a CCA-based method supplemented by the essence of LDA is proposed for semi-supervised feature extraction from multiview data.KeywordsSemisupervised LearningFeature ExtractionMultiview LearningLDACCA

Dimensionality reduction of image features using the canonical contextual correlation projection

Many of the most widely accepted methods for reliable detection of steady-state visual evoked potentials (SSVEPs) in the electroencephalogram (EEG) utilize canonical correlation analysis (CCA). CCA uses pure sine and cosine reference templates with frequencies corresponding to the visual stimulation frequencies. These generic reference templates may not optimally reflect the natural SSVEP features obscured by the background EEG. This paper introduces a new approach that utilizes spatio-temporal feature extraction with multivariate linear regression (MLR) to learn discriminative SSVEP features for improving the detection accuracy. MLR is implemented on dimensionality-reduced EEG training data and a constructed label matrix to find optimally discriminative subspaces. Experimental results show that the proposed MLR method significantly outperforms CCA as well as several other competing methods for SSVEP detection, especially for time windows shorter than 1 second. This demonstrates that the MLR method is a promising new approach for achieving improved real-time performance of SSVEP-BCIs.

Ambiguity function(AF) modeling of radar signals is a powerful approach to feature extraction and recognition of radar emitters.An AF subspace based optimization framework is proposed to identify radar emitters by exploring unintentional modulation on pulse(UMOP) features.First,near-zero Doppler cuts of AF were extracted as a preliminary feature subset.Then,two kinds of cut-concatenation schemes were designed to construct two different pairs of feature vectors with complementary information respectively,which will facilitate the subsequent feature fusion via canonical correlation analysis(CCA) or discriminative canonical correlation analysis(DCCA).Theoretical analysis and experimental results show that the proposed algorithms not only alleviate the calculation problem in the existing AF based method,but also improve the recognition performance considerably,due to the successful information fusion and redundancy reduction conducted in the AF subset.

A feature extraction method for Facial Expression Recognition Systems is proposed based on CCA using FMEDA and DPA. For proper classification of expression it has been trained with Recurrent neural network. The Cohn Kanade Extensive, ORL and Yale databases are used in this paper. The images have been preprocessed using image normalization and then contrast limited adaptive histogram equalization to remove the illumination variance and noises. After down-sampling, the dimensions with factor data is provided to Canonical Correlation Analysis that finds the linear combinations among two sets of variables for determining correlation with each other. DPA and FMEDA have been used to extract features from specially low frequency coefficients of the image as some of the low frequency coefficients have more discrimination power compared to others and by extracting those features a higher true recognition rate can be achieved. The results provided in this paper show the advantage of this method compared to other methods like Principal Component Analysis and Linear Discriminant Analysis which uses between and within class scatter matrices and try to maximize the discrimination in transformed domain. This method searches for best discrimination features in transformed domain. Dimensionally reduced data is provided to recurrent neural network for training purpose. Meanwhile, the proposed method is more robust and effective compared to other methods in this field.

Steady State Visual Evoked Potential (SSVEP) has been commonly adopted in Brain Computer Interface (BCI) applications. For wearable BCI applications, several aspects of SSVEP-based BCI systems, such as speed, subject variability, and accurate target detection, are under ongoing research investigations. Up to date, Canonical Correlation Analysis (CCA) has been considered the state-of-the-art feature extraction method for SSVEP-based BCI systems. Nevertheless, although CCA outperforms traditional SSVEP detection methods, such as Power Spectral Density Analysis (PSDA), achieving high accuracies when detecting target frequencies is still a challenging task due to user variation and physiological changes in the human body. In this paper, we investigate an SSVEP-based BCI application using wireless EEG recording and an Android tablet-based user interface. We propose a fusion of CCA and PSDA solutions at the score level by dividing their score space into multiple partitions, and extract and combine their complementary discriminative information to minimize the detection error in a linear fashion. We investigated transforming the fusion score space to lower dimensions with the purpose of alleviating redundancy. As such, we employed Principal Component Analysis (PCA), and Linear Discriminant Analysis (LDA). Our experimental results demonstrated that our proposed score fusion method is effective in reducing the effect of noise and non-stationary elements in EEG dynamics. Average detection accuracies improved from 63% for CCA to 72% for fusion+PCA and further improved to 98% for fusion+LDA.

Video event carries very rich and complex semantics, and video event analysis is intrinsically multi-view learning problem. In this paper, by revealing the relationship between canonical correlation analysis (CCA) and linear discriminant analysis (LDA), we propose a new convenient multi-view learning architecture, i.e., multi-view discriminative fusion on CCA (MvDF-CCA). Instead of pair-wise relation consideration among different views, MvDF-CCA separately aligns each view space to the target one comprised by labeled indicators, which is easy expanding to additional view data. MvDF-CCA leverages the discriminative ability in LDA and the strengths in CCA, and maps all view spaces to the same target labeled space, which can be concurrent execution on large scale video categorization. Competitive results are reported on the well-known UCF101 and Columbia Consumer Video (CCV) benchmarks.

Spectrum-Enhanced TRCA (SE-TRCA): A novel approach for direction detection in SSVEP-based BCI

We address the problem of comparing sets of images for object recognition, where the sets may represent variations in an object's appearance due to changing camera pose and lighting conditions. Canonical Correlations (also known as principal or canonical angles), which can be thought of as the angles between two d-dimensional subspaces, have recently attracted attention for image set matching. Canonical correlations offer many benefits in accuracy, efficiency, and robustness compared to the two main classical methods: parametric distribution-based and nonparametric sample-based matching of sets. Here, this is first demonstrated experimentally for reasonably sized data sets using existing methods exploiting canonical correlations. Motivated by their proven effectiveness, a novel discriminative learning method over sets is proposed for set classification. Specifically, inspired by classical Linear Discriminant Analysis (LDA), we develop a linear discriminant function that maximizes the canonical correlations of within-class sets and minimizes the canonical correlations of between-class sets. Image sets transformed by the discriminant function are then compared by the canonical correlations. Classical orthogonal subspace method (OSM) is also investigated for the similar purpose and compared with the proposed method. The proposed method is evaluated on various object recognition problems using face image sets with arbitrary motion captured under different illuminations and image sets of 500 general objects taken at different views. The method is also applied to object category recognition using ETH-80 database. The proposed method is shown to outperform the state-of-the-art methods in terms of accuracy and efficiency.

Feature fusion is important for providing enhancements of data authenticity in both traditional and deep learning pattern analysis. Classical serial fusion concatenates multiple feature sets followed by dimensionality reduction using principal component analysis (PCA), linear discriminant analysis (LDA), canonical correlation analysis (CCA) etc. CCA-based feature fusion is a main technique for exploring the mutual relationships of multiple feature sets. It considers the correlation of multiple feature sets during dimensionality reduction. In traditional CCA-based feature fusion and extraction, the natural groupings of features are directly used. It is still unclear whether the natural groupings of features are optimal for CCA-based fusion. In this paper, we propose a feature regrouping algorithm for CCA-based feature fusion and extraction through normalized cut (FR-NC). Feature correlation analysis is incorporated into normalized cut, in which the intra-group correlation is maximized, and the extra-group correlation is minimized simultaneously. CCA-based feature fusion is performed on the regrouped features. The proposed feature regrouping algorithm aims to provide enhanced fused features for pattern classification. Extensive experiments have proved its effectiveness.

Feature extraction for classification problems and its application to face recognition

Chapter 11 - Canonical analysis

Latent Linear Discriminant Analysis for feature extraction via Isometric Structural Learning

M25 - Investigation of Relationships Between Bipolar Disorder Phenotypes And Genome-Wide Significant Loci From PGC2 Schizophrenia