General Tensor Discriminant Analysis Research Articles

Tensor-Based Emotional Category Classification via Visual Attention-Based Heterogeneous CNN Feature Fusion.

The paper proposes a method of visual attention-based emotion classification through eye gaze analysis. Concretely, tensor-based emotional category classification via visual attention-based heterogeneous convolutional neural network (CNN) feature fusion is proposed. Based on the relationship between human emotions and changes in visual attention with time, the proposed method performs new gaze-based image representation that is suitable for reflecting the characteristics of the changes in visual attention with time. Furthermore, since emotions evoked in humans are closely related to objects in images, our method uses a CNN model to obtain CNN features that can represent their characteristics. For improving the representation ability to the emotional categories, we extract multiple CNN features from our novel gaze-based image representation and enable their fusion by constructing a novel tensor consisting of these CNN features. Thus, this tensor construction realizes the visual attention-based heterogeneous CNN feature fusion. This is the main contribution of this paper. Finally, by applying logistic tensor regression with general tensor discriminant analysis to the newly constructed tensor, the emotional category classification becomes feasible. Since experimental results show that the proposed method enables the emotional category classification with the F1-measure of approximately 0.6, and about 10% improvement can be realized compared to comparative methods including state-of-the-art methods, the effectiveness of the proposed method is verified.

Selection of Significant Brain Regions Based on MvGTDA and TS-DLF for Emotion Estimation

In this paper, we propose a novel method for estimating human emotion using functional brain images. The final goal of our study is contribution to affective brain computer interfaces (aBCIs), which use neuropsychological signals. In the proposed method, we newly derive multiview general tensor discriminant analysis (MvGTDA) in order to reveal significant brain regions and accurately estimate human emotion evoked by visual stimuli. This is because it is important to find activation of multiple brain regions for estimating emotional states. Since we regard a “Brodmann area” as a “view” and introduce $L_{1}$ -norm regularization for these views, MvGTDA can eliminate non-crucial Brodmann areas and select significant ones. Moreover, in general studies on functional brain images based on machine learning methodologies, there is an overfitting problem caused by a small sample size. Therefore, revealing significant Brodmann areas based on MvGTDA has another important role, i.e., solving the overfitting problem. By inputting estimation results respectively obtained from the significant areas and the MvGTDA-based feature, tensor-based supervised decision-level fusion (TS-DLF) integrates them and outputs the final estimation result of the user’s emotion. In experiments, we showed the effectiveness of our method by using actual functional brain images and we revealed the significant brain regions in emotional states.

Optimized Seizure Detection Algorithm: A Fast Approach for Onset of Epileptic in EEG Signals Using GT Discriminant Analysis and K-NN Classifier.

Epilepsy is a severe disorder of the central nervous system that predisposes the person to recurrent seizures. Fifty million people worldwide suffer from epilepsy; after Alzheimer's and stroke, it is the third widespread nervous disorder. In this paper, an algorithm to detect the onset of epileptic seizures based on the analysis of brain electrical signals (EEG) has been proposed. 844 hours of EEG were recorded form 23 pediatric patients consecutively with 163 occurrences of seizures. Signals had been collected from Children's Hospital Boston with a sampling frequency of 256 Hz through 18 channels in order to assess epilepsy surgery. By selecting effective features from seizure and non-seizure signals of each individual and putting them into two categories, the proposed algorithm detects the onset of seizures quickly and with high sensitivity. In this algorithm, L-sec epochs of signals are displayed in form of a third-order tensor in spatial, spectral and temporal spaces by applying wavelet transform. Then, after applying general tensor discriminant analysis (GTDA) on tensors and calculating mapping matrix, feature vectors are extracted. GTDA increases the sensitivity of the algorithm by storing data without deleting them. Finally, K-Nearest neighbors (KNN) is used to classify the selected features. The results of simulating algorithm on algorithm standard dataset shows that the algorithm is capable of detecting 98 percent of seizures with an average delay of 4.7 seconds and the average error rate detection of three errors in 24 hours. Today, the lack of an automated system to detect or predict the seizure onset is strongly felt.

A Report on Multilinear PCA Plus GTDA to Deal With Face Image

Abstract Because face images are naturally two-dimensional data, there have been several 2D feature extraction methods to deal with facial images while there are few 2D effective classifiers. Meanwhile, there is an increasing interest in the multilinear subspace analysis and many methods have been proposed to operate directly on these tensorial data during the past several years. One of these popular unsupervised multilinear algorithms is Multilinear Principal Component Analysis (MPCA) while another of the supervised multilinear algorithm is Multilinear Discriminant Analysis (MDA). Then a MPCA+MDA method has been introduced to deal with the tensorial signal. However, due to the no convergence of MDA, it is difficult for MPCA+MDA to obtain a precise result. Hence, to overcome this limitation, a new MPCA plus General Tensor Discriminant Analysis (GTDA) solution with well convergence is presented for tensorial face images feature extraction in this paper. Several experiments are carried out to evaluate the performance of MPCA+GTDA on different databases and the results show that this method has the potential to achieve comparative effect as MPCA+MDA.

Multilinear Discriminant Analysis for Higher-Order Tensor Data Classification.

In the past decade, great efforts have been made to extend linear discriminant analysis for higher-order data classification, generally referred to as multilinear discriminant analysis (MDA). Existing examples include general tensor discriminant analysis (GTDA) and discriminant analysis with tensor representation (DATER). Both the two methods attempt to resolve the problem of tensor mode dependency by iterative approximation. GTDA is known to be the first MDA method that converges over iterations. However, its performance relies highly on the tuning of the parameter in the scatter difference criterion. Although DATER usually results in better classification performance, it does not converge, yet the number of iterations executed has a direct impact on DATER's performance. In this paper, we propose a closed-form solution to the scatter difference objective in GTDA, namely, direct GTDA (DGTDA) which also gets rid of parameter tuning. We demonstrate that DGTDA outperforms GTDA in terms of both efficiency and accuracy. In addition, we propose constrained multilinear discriminant analysis (CMDA) that learns the optimal tensor subspace by iteratively maximizing the scatter ratio criterion. We prove both theoretically and experimentally that the value of the scatter ratio criterion in CMDA approaches its extreme value, if it exists, with bounded error, leading to superior and more stable performance in comparison to DATER.

A Novel Fast Epileptic Seizure Onset Detection Algorithm Using General Tensor Discriminant Analysis

Seizure onset detection with minimum latency has a key role in improving the therapy studies of epilepsy. In this article, an epileptic seizure onset detection algorithm based on general tensor discriminant analysis is proposed to detect the seizure through EEG signals with smallest delay before the development of clinical symptoms. In this algorithm, seizure and nonseizure EEG signal epochs are exhibited by spectral, spatial, and temporal domains (third-order tensors) in wavelet decomposition. Then, to reduce feature space, projection matrices are extracted from tensor-represented EEG signal by general tensor discriminant analysis. In this strategy, the discriminative information in the training tensors is preserved that it is a benefit in comparison with common feature space reduction algorithms such as principal component analysis and multilinear subspace analysis. The proposed seizure onset detection algorithm is evaluated on 44 epileptic patients from 2 standard datasets and recognizes 98% of seizures with average delay of 4.5 seconds. The obtained results show efficiency and effectiveness of our proposed algorithm in comparison with other algorithms.

Discriminant Subspace Learning for Microcalcification Clusters Detection

This paper presents a novel approach to microcalcification clusters (MCs) detection in mammograms based on the discriminant subspace learning. The ground truth of MCs in mammograms is assumed to be known as a priori. Several typical subspace learning algorithms, such as principal component analysis (PCA), linear discriminant analysis (LDA), tensor subspace analysis (TSA) and general tensor discriminant Analysis (GTDA), are employed to extract subspace features. In subspace feature domain, the MCs detection procedure is formulated as a supervised learning and classification problem, and SVM is used as a classifier to make decision for the presence of MCs or not. A large number of experiments are carried out to evaluate and compare the performance of the proposed MCs detection algorithms. The experiment result suggests that correlation filters is a promising technique for MCs detection.

A Prior Neurophysiologic Knowledge Free Tensor-Based Scheme for Single Trial EEG Classification

Single trial electroencephalogram (EEG) classification is essential in developing brain-computer interfaces (BCIs). However, popular classification algorithms, e.g., common spatial patterns (CSP), usually highly depend on the prior neurophysiologic knowledge for noise removing, although this knowledge is not always known in practical applications. In this paper, a novel tensor-based scheme is proposed for single trial EEG classification, which performs well without the prior neurophysiologic knowledge. In this scheme, EEG signals are represented in the spatial-spectral-temporal domain by the wavelet transform, the multilinear discriminative subspace is reserved by the general tensor discriminant analysis (GTDA), redundant indiscriminative patterns are removed by Fisher score, and the classification is conducted by the support vector machine (SVM). Applications to three datasets confirm the effectiveness and the robustness of the proposed tensor scheme in analyzing EEG signals, especially in the case of lacking prior neurophysiologic knowledge.

Discriminative optical flow tensor for video semantic analysis

This paper presents a novel framework for effective video semantic analysis. This framework has two major components, namely, optical flow tensor (OFT) and hidden Markov models (HMMs). OFT and HMMs are employed because: (1) motion is one of the fundamental characteristics reflecting the semantic information in video, so an OFT-based feature extraction method is developed to make full use of the motion information. Thereafter, to preserve the structure and discriminative information presented by OFT, general tensor discriminant analysis (GTDA) is used for dimensionality reduction. Finally, linear discriminant analysis (LDA) is utilized to further reduce the feature dimension for discriminative motion information representation; and (2) video is a sort of information intensive sequential media characterized by its context-sensitive nature, so the video sequences can be more effectively analyzed by some temporal modeling tools. In this framework, we use HMMs to well model different levels of semantic units (SU), e.g., shot and event. Experimental results are reported to demonstrate the advantages of the proposed framework upon semantic analysis of basketball video sequences, and the cross validations illustrate its feasibility and effectiveness.

General Tensor Discriminant Analysis and Gabor Features for Gait Recognition

The traditional image representations are not suited to conventional classification methods, such as the linear discriminant analysis (LDA), because of the under sample problem (USP): the dimensionality of the feature space is much higher than the number of training samples. Motivated by the successes of the two dimensional LDA (2DLDA) for face recognition, we develop a general tensor discriminant analysis (GTDA) as a preprocessing step for LDA. The benefits of GTDA compared with existing preprocessing methods, e.g., principal component analysis (PCA) and 2DLDA, include 1) the USP is reduced in subsequent classification by, for example, LDA; 2) the discriminative information in the training tensors is preserved; and 3) GTDA provides stable recognition rates because the alternating projection optimization algorithm to obtain a solution of GTDA converges, while that of 2DLDA does not. We use human gait recognition to validate the proposed GTDA. The averaged gait images are utilized for gait representation. Given the popularity of Gabor function based image decompositions for image understanding and object recognition, we develop three different Gabor function based image representations: 1) the GaborD representation is the sum of Gabor filter responses over directions, 2) GaborS is the sum of Gabor filter responses over scales, and 3) GaborSD is the sum of Gabor filter responses over scales and directions. The GaborD, GaborS and GaborSD representations are applied to the problem of recognizing people from their averaged gait images.A large number of experiments were carried out to evaluate the effectiveness (recognition rate) of gait recognition based on first obtaining a Gabor, GaborD, GaborS or GaborSD image representation, then using GDTA to extract features and finally using LDA for classification. The proposed methods achieved good performance for gait recognition based on image sequences from the USF HumanID Database. Experimental comparisons are made with nine state of the art classification methods in gait recognition.