scispace - formally typeset
Search or ask a question

Showing papers on "Linear discriminant analysis published in 2015"


Journal ArticleDOI
TL;DR: Surprisingly, for all tasks, such a seemingly naive PCANet model is on par with the state-of-the-art features either prefixed, highly hand-crafted, or carefully learned [by deep neural networks (DNNs)].
Abstract: In this paper, we propose a very simple deep learning network for image classification that is based on very basic data processing components: 1) cascaded principal component analysis (PCA); 2) binary hashing; and 3) blockwise histograms. In the proposed architecture, the PCA is employed to learn multistage filter banks. This is followed by simple binary hashing and block histograms for indexing and pooling. This architecture is thus called the PCA network (PCANet) and can be extremely easily and efficiently designed and learned. For comparison and to provide a better understanding, we also introduce and study two simple variations of PCANet: 1) RandNet and 2) LDANet. They share the same topology as PCANet, but their cascaded filters are either randomly selected or learned from linear discriminant analysis. We have extensively tested these basic networks on many benchmark visual data sets for different tasks, including Labeled Faces in the Wild (LFW) for face verification; the MultiPIE, Extended Yale B, AR, Facial Recognition Technology (FERET) data sets for face recognition; and MNIST for hand-written digit recognition. Surprisingly, for all tasks, such a seemingly naive PCANet model is on par with the state-of-the-art features either prefixed, highly hand-crafted, or carefully learned [by deep neural networks (DNNs)]. Even more surprisingly, the model sets new records for many classification tasks on the Extended Yale B, AR, and FERET data sets and on MNIST variations. Additional experiments on other public data sets also demonstrate the potential of PCANet to serve as a simple but highly competitive baseline for texture classification and object recognition.

1,034 citations


Book
01 Jan 2015
TL;DR: Cumming et al. as discussed by the authors presented a review of basic statistics for SPSS and other useful procedures, including the following: 1. Introduction 2. Data Coding and Exploratory Analysis (EDA) 3. Imputation of Missing Data 4. Several Measures of Reliability 5. Selecting and Interpreting Inferential Statistics 7. Multiple Regression 8. Mediation, Moderation, and Canonical Correlation 9. Logistic Regression and Discriminant Analysis 10. Factorial ANOVA and ANCOVA 11. Repeated-Me
Abstract: 1. Introduction 2. Data Coding and Exploratory Analysis (EDA) 3. Imputation of Missing Data 4. Several Measures of Reliability 5. Exploratory Factor Analysis and Principal Components Analysis 6. Selecting and Interpreting Inferential Statistics 7. Multiple Regression 8. Mediation, Moderation, and Canonical Correlation 9. Logistic Regression and Discriminant Analysis 10. Factorial ANOVA and ANCOVA 11. Repeated-Measures and Mixed ANOVAs 12. Multivariate Analysis of Variance (MANOVA) 13. Multilevel Linear Modeling/Hierarchical Linear Modeling Appendix A. Getting Started With SPSS and Other Useful Procedures D. Quick, M. Myers Appendix B. Review of Basic Statistics J.M. Cumming, A. Weinberg Appendix C. Answers to Odd Interpretation Questions

854 citations


Journal ArticleDOI
TL;DR: This tutorial review aims to provide an introductory overview to several straightforward statistical methods such as principal component-discriminant function analysis (PC-DFA), support vector machines (SVM) and random forests (RF), which could very easily be used either to augment PLS or as alternative supervised learning methods to PLS-DA.

606 citations


Journal ArticleDOI
TL;DR: A comparison of traditional statistical and novel machine learning models applied for regional scale landslide susceptibility modeling is presented and it is suggested that the framework of this model evaluation approach can be applied to assist in selection of a suitable landslide susceptibility modeled technique.

515 citations


Journal Article
TL;DR: This survey and generic solver suggest that linear dimensionality reduction can move toward becoming a blackbox, objective-agnostic numerical technology.
Abstract: Linear dimensionality reduction methods are a cornerstone of analyzing high dimensional data, due to their simple geometric interpretations and typically attractive computational properties. These methods capture many data features of interest, such as covariance, dynamical structure, correlation between data sets, input-output relationships, and margin between data classes. Methods have been developed with a variety of names and motivations in many fields, and perhaps as a result the connections between all these methods have not been highlighted. Here we survey methods from this disparate literature as optimization programs over matrix manifolds. We discuss principal component analysis, factor analysis, linear multidimensional scaling, Fisher's linear discriminant analysis, canonical correlations analysis, maximum autocorrelation factors, slow feature analysis, sufficient dimensionality reduction, undercomplete independent component analysis, linear regression, distance metric learning, and more. This optimization framework gives insight to some rarely discussed shortcomings of well-known methods, such as the suboptimality of certain eigenvector solutions. Modern techniques for optimization over matrix manifolds enable a generic linear dimensionality reduction solver, which accepts as input data and an objective to be optimized, and returns, as output, an optimal low-dimensional projection of the data. This simple optimization framework further allows straightforward generalizations and novel variants of classical methods, which we demonstrate here by creating an orthogonal-projection canonical correlations analysis. More broadly, this survey and generic solver suggest that linear dimensionality reduction can move toward becoming a blackbox, objective-agnostic numerical technology.

430 citations


Journal ArticleDOI
TL;DR: In this paper, the authors define a jet image using calorimeter towers as the elements of the image and establish jet-image preprocessing methods, and develop a discriminant for classifying the jet-images derived using Fisher discriminant analysis.
Abstract: We introduce a novel approach to jet tagging and classification through the use of techniques inspired by computer vision. Drawing parallels to the problem of facial recognition in images, we define a jet-image using calorimeter towers as the elements of the image and establish jet-image preprocessing methods. For the jet-image processing step, we develop a discriminant for classifying the jet-images derived using Fisher discriminant analysis. The effectiveness of the technique is shown within the context of identifying boosted hadronic W boson decays with respect to a background of quark- and gluoninitiated jets. Using Monte Carlo simulation, we demonstrate that the performance of this technique introduces additional discriminating power over other substructure approaches, and gives significant insight into the internal structure of jets.

237 citations


Journal ArticleDOI
TL;DR: A marginal feature screening procedure based on empirical conditional distribution function is proposed that is model-free in that its implementation does not require specification of a regression model and robust to heavy-tailed distributions of predictors and the presence of potential outliers.
Abstract: This work is concerned with marginal sure independence feature screening for ultrahigh dimensional discriminant analysis. The response variable is categorical in discriminant analysis. This enables us to use the conditional distribution function to construct a new index for feature screening. In this article, we propose a marginal feature screening procedure based on empirical conditional distribution function. We establish the sure screening and ranking consistency properties for the proposed procedure without assuming any moment condition on the predictors. The proposed procedure enjoys several appealing merits. First, it is model-free in that its implementation does not require specification of a regression model. Second, it is robust to heavy-tailed distributions of predictors and the presence of potential outliers. Third, it allows the categorical response having a diverging number of classes in the order of O(nκ) with some κ ⩾ 0. We assess the finite sample property of the proposed procedure by Mont...

194 citations


Journal ArticleDOI
Yuan Liu1, Yanmin Qian1, Nanxin Chen1, Tianfan Fu1, Ya Zhang1, Kai Yu1 
TL;DR: Experiments showed that deep feature based methods can obtain significant performance improvements compared to the traditional baselines, no matter if they are directly applied in the GMM-UBM system or utilized as identity vectors.

176 citations


Journal ArticleDOI
TL;DR: An accurate and robust facial expression recognition (FER) system that employs stepwise linear discriminant analysis (SWLDA), which is a significant improvement in contrast to the existing FER methods.
Abstract: This paper introduces an accurate and robust facial expression recognition (FER) system. For feature extraction, the proposed FER system employs stepwise linear discriminant analysis (SWLDA). SWLDA focuses on selecting the localized features from the expression frames using the partial $\boldsymbol {F}$ -test values, thereby reducing the within class variance and increasing the low between variance among different expression classes. For recognition, the hidden conditional random fields (HCRFs) model is utilized. HCRF is capable of approximating a complex distribution using a mixture of Gaussian density functions. To achieve optimum results, the system employs a hierarchical recognition strategy. Under these settings, expressions are divided into three categories based on parts of the face that contribute most toward an expression. During recognition, at the first level, SWLDA and HCRF are employed to recognize the expression category; whereas, at the second level, the label for the expression within the recognized category is determined using a separate set of SWLDA and HCRF, trained just for that category. In order to validate the system, four publicly available data sets were used, and a total of four experiments were performed. The weighted average recognition rate for the proposed FER approach was 96.37% across the four different data sets, which is a significant improvement in contrast to the existing FER methods.

159 citations


Journal ArticleDOI
TL;DR: Six supervised learning algorithms, including linear discriminant analysis, multinomial logistic regression, multilayer perceptron neural networks, support vector machine (SVM), random forest (RF), and gradient boosting machine are evaluated for their ability to learn for PS based on different input parameter combinations.
Abstract: The prediction of pillar stability (PS) in hard rock mines is a crucial task for which many techniques and methods have been proposed in the literature including machine learning classification. In order to make the best use of the large variety of statistical and machine learning classification methods available, it is necessary to assess their performance before selecting a classifier and suggesting improvement. The objective of this paper is to compare different classification techniques for PS detection in hard rock mines. The data of this study consist of six features, namely pillar width, pillar height, the ratio of pillar width to its height, uniaxial compressive strength of the rock, pillar strength, and pillar stress. A total of 251 pillar cases between 1972 and 2011 are analyzed. Six supervised learning algorithms, including linear discriminant analysis, multinomial logistic regression, multilayer perceptron neural networks, support vector machine (SVM), random forest (RF), and gradient boosting machine, are evaluated for their ability to learn for PS based on different input parameter combinations. In this study, the available data set is randomly split into two parts: training set (70 %) and test set (30 %). A repeated tenfold cross-validation procedure (ten repeats) is applied to determine the optimal parameter values during modeling, and an external testing set is employed to validate the prediction performance of models. Two performance measures, namely classification accuracy rate and Cohen’s kappa, are employed. The analysis of the accuracy together with kappa for the PS data set demonstrates that SVM and RF achieve comparable median classification accuracy rate and Cohen’s kappa values. All models are fitted by “R” programs with the libraries and functions described in this study.

155 citations


Journal ArticleDOI
TL;DR: A classification approach that hybridizes statistical techniques and SOM for network anomaly detection and Probabilistic Self-Organizing Maps (PSOM) aim to model the feature space and enable distinguishing between normal and anomalous connections.

Posted Content
TL;DR: A unified analysis of the predictive risk of ridge regression and regularized discriminant analysis in a dense random effects model in a high-dimensional asymptotic regime and finds that predictive accuracy has a nuanced dependence on the eigenvalue distribution of the covariance matrix.
Abstract: We provide a unified analysis of the predictive risk of ridge regression and regularized discriminant analysis in a dense random effects model. We work in a high-dimensional asymptotic regime where $p, n \to \infty$ and $p/n \to \gamma \in (0, \, \infty)$, and allow for arbitrary covariance among the features. For both methods, we provide an explicit and efficiently computable expression for the limiting predictive risk, which depends only on the spectrum of the feature-covariance matrix, the signal strength, and the aspect ratio $\gamma$. Especially in the case of regularized discriminant analysis, we find that predictive accuracy has a nuanced dependence on the eigenvalue distribution of the covariance matrix, suggesting that analyses based on the operator norm of the covariance matrix may not be sharp. Our results also uncover several qualitative insights about both methods: for example, with ridge regression, there is an exact inverse relation between the limiting predictive risk and the limiting estimation risk given a fixed signal strength. Our analysis builds on recent advances in random matrix theory.

Journal ArticleDOI
TL;DR: A Modified Fisher Discriminant Function is proposed in this study which makes the traditional function more sensitive to the important instances, so that the profit that can be obtained from a fraud/legitimate classifier is maximized.
Abstract: We introduce Fisher Linear Discriminant Analysis (FLDA).We modify it to be sensitive toward profitable instances.We applied them together in credit card fraud detection problem.The results are compared in terms of total obtained profit with three well-known models.Modified fisher outperforms the other models in attaining high profit. In parallel to the increase in the number of credit card transactions, the financial losses due to fraud have also increased. Thus, the popularity of credit card fraud detection has been increased both for academicians and banks. Many supervised learning methods were introduced in credit card fraud literature some of which bears quite complex algorithms. As compared to complex algorithms which somehow over-fit the dataset they are built on, one can expect simpler algorithms may show a more robust performance on a range of datasets. Although, linear discriminant functions are less complex classifiers and can work on high-dimensional problems like credit card fraud detection, they did not receive considerable attention so far. This study investigates a linear discriminant, called Fisher Discriminant Function for the first time in credit card fraud detection problem. On the other hand, in this and some other domains, cost of false negatives is very higher than false positives and is different for each transaction. Thus, it is necessary to develop classification methods which are biased toward the most important instances. To cope for this, a Modified Fisher Discriminant Function is proposed in this study which makes the traditional function more sensitive to the important instances. This way, the profit that can be obtained from a fraud/legitimate classifier is maximized. Experimental results confirm that Modified Fisher Discriminant could eventuate more profit.

Journal ArticleDOI
TL;DR: In this article, the authors present an overview of linear discriminant analysis (LDA) techniques for solving small sample size (SSS) problem and highlight some important datasets and software/packages.
Abstract: Dimensionality reduction is an important aspect in the pattern classification literature, and linear discriminant analysis (LDA) is one of the most widely studied dimensionality reduction technique. The application of variants of LDA technique for solving small sample size (SSS) problem can be found in many research areas e.g. face recognition, bioinformatics, text recognition, etc. The improvement of the performance of variants of LDA technique has great potential in various fields of research. In this paper, we present an overview of these methods. We covered the type, characteristics and taxonomy of these methods which can overcome SSS problem. We have also highlighted some important datasets and software/packages.

Proceedings ArticleDOI
11 May 2015
TL;DR: A systematic approach for the automated training of condition monitoring systems for complex hydraulic systems is developed and evaluated and the classification rate for random load cycles was enhanced by a distribution analysis of feature trends.
Abstract: In this paper, a systematic approach for the automated training of condition monitoring systems for complex hydraulic systems is developed and evaluated. We analyzed different fault scenarios using a test rig that allows simulating a reversible degradation of component's conditions. By analyzing the correlation of features extracted from raw sensor data and the known fault characteristics of experimental obtained data, the most significant features specific to a fault case can be identified. These feature values are transferred to a lower-dimensional discriminant space using linear discriminant analysis (LDA) which allows the classification of fault condition and grade of severity. We successfully implemented and tested the system for a fixed working cycle of the hydraulic system. Furthermore, the classification rate for random load cycles was enhanced by a distribution analysis of feature trends.

01 Jan 2015
TL;DR: The type, characteristics and taxonomy of these methods which can overcome SSS problem are covered and some important datasets and software/packages are highlighted.

Journal ArticleDOI
TL;DR: A novel clustering algorithm with stimulating discriminant measures (SDM) of both within- and between-cluster scatter variances is proposed to produce robust segmentation of nucleus and cytoplasm of lymphocytes/lymphoblasts for acute lymphoblastic leukaemia diagnosis from microscopic blood images.
Abstract: This research proposes an intelligent decision support system for acute lymphoblastic leukaemia diagnosis from microscopic blood images. A novel clustering algorithm with stimulating discriminant measures (SDM) of both within- and between-cluster scatter variances is proposed to produce robust segmentation of nucleus and cytoplasm of lymphocytes/lymphoblasts. Specifically, the proposed between-cluster evaluation is formulated based on the trade-off of several between-cluster measures of well-known feature extraction methods. The SDM measures are used in conjuction with Genetic Algorithm for clustering nucleus, cytoplasm, and background regions. Subsequently, a total of eighty features consisting of shape, texture, and colour information of the nucleus and cytoplasm sub-images are extracted. A number of classifiers (multi-layer perceptron, Support Vector Machine (SVM) and Dempster-Shafer ensemble) are employed for lymphocyte/lymphoblast classification. Evaluated with the ALL-IDB2 database, the proposed SDM-based clustering overcomes the shortcomings of Fuzzy C-means which focuses purely on within-cluster scatter variance. It also outperforms Linear Discriminant Analysis and Fuzzy Compactness and Separation for nucleus-cytoplasm separation. The overall system achieves superior recognition rates of 96.72% and 96.67% accuracies using bootstrapping and 10-fold cross validation with Dempster-Shafer and SVM, respectively. The results also compare favourably with those reported in the literature, indicating the usefulness of the proposed SDM-based clustering method.

Journal ArticleDOI
TL;DR: This paper achieves automated artifact elimination using linear discriminant analysis (LDA) for classification of feature vectors extracted from ICA components via image processing algorithms and identifies range filtering as a feature extraction method with great potential for automated IC artifact recognition.

Journal ArticleDOI
TL;DR: An automated multi-class classification approach for tomato ripeness measurement and evaluation via investigating and classifying the different maturity/ripeness stages using Principal Components Analysis (PCA) in addition to Support Vector Machines (SVMs) and Linear Discriminant Analysis (LDA) algorithms for feature extraction and classification, respectively.
Abstract: Egypt occupied the fifth place in both income and weight of tomato production.We proposed an automated multi-class classification approach for tomato ripeness stages.Performance of classification algorithms depends on statistics of the experimented dataset.Training and testing datasets have been generated via employing the 10-fold cross validation.Using OAO multi-class SVMs with linear kernel function outperformed other algorithms. Tomato quality is one of the most important factors that helps ensuring a consistent marketing of tomato fruit. As ripeness is the main indicator for tomato quality from customers perspective, the determination of tomato ripeness stages is a basic industrial concern regarding tomato production in order to get high quality product. Automatic ripeness evaluation of tomato is an essential research topic as it may prove benefits in ensuring optimum yield of high quality product, this will increase the income because tomato is one of the most important crops in the world. This article presents an automated multi-class classification approach for tomato ripeness measurement and evaluation via investigating and classifying the different maturity/ripeness stages. The proposed approach uses color features for classifying tomato ripeness stages. The approach proposed in this article uses Principal Components Analysis (PCA) in addition to Support Vector Machines (SVMs) and Linear Discriminant Analysis (LDA) algorithms for feature extraction and classification, respectively. Experiments have been conducted on a dataset of total 250 images that has been used for both training and testing datasets with 10-fold cross validation. Experimental results showed that the proposed classification approach has obtained ripeness classification accuracy of 90.80%, using one-against-one (OAO) multi-class SVMs algorithm with linear kernel function, ripeness classification accuracy of 84.80% using one-against-all (OAA) multi-class SVMs algorithm with linear kernel function, and ripeness classification accuracy of 84% using LDA algorithm.

Proceedings ArticleDOI
Wen Wang1, Ruiping Wang1, Zhiwu Huang1, Shiguang Shan1, Xilin Chen1 
07 Jun 2015
TL;DR: The proposed method, Discriminant Analysis on Riemannian manifold of Gaussian distributions (DARG), is evaluated by face identification and verification tasks on four most challenging and largest databases, YouTube Celebrities, COX, YouTube Face DB and Point-and-Shoot Challenge, to demonstrate its superiority over the state-of-the-art.
Abstract: This paper presents a method named Discriminant Analysis on Riemannian manifold of Gaussian distributions (DARG) to solve the problem of face recognition with image sets. Our goal is to capture the underlying data distribution in each set and thus facilitate more robust classification. To this end, we represent image set as Gaussian Mixture Model (GMM) comprising a number of Gaussian components with prior probabilities and seek to discriminate Gaussian components from different classes. In the light of information geometry, the Gaussians lie on a specific Riemannian manifold. To encode such Riemannian geometry properly, we investigate several distances between Gaussians and further derive a series of provably positive definite probabilistic kernels. Through these kernels, a weighted Kernel Discriminant Analysis is finally devised which treats the Gaussians in GMMs as samples and their prior probabilities as sample weights. The proposed method is evaluated by face identification and verification tasks on four most challenging and largest databases, YouTube Celebrities, COX, YouTube Face DB and Point-and-Shoot Challenge, to demonstrate its superiority over the state-of-the-art.

Journal ArticleDOI
TL;DR: This paper proposes an L1-norm two-dimensional linear discriminant analysis (L1-2DLDA) with robust performance, which is supported by the preliminary experiments on toy example and face datasets, which show the improvement of the L 1-2 DLDA over L2- 2DLDA.

Journal ArticleDOI
TL;DR: A novel method including segmentation, a combination of new and well-known feature extraction and classification methods to classify plant leaves to distinguish leaf margins, which cannot be distinguished using commonly used geometric features.

Journal ArticleDOI
TL;DR: Different morphological and texture features widely used in computer-aided diagnosis systems for BUS images are compiled and evaluated for classifying breast lesions on ultrasound, revealing the best classification performance is obtained by a morphological set with five features.

Journal ArticleDOI
TL;DR: An overview of the model-based clustering and classification methods implemented in Mixmod is given, and it is shown how the R package Rmixmod can be used for clustersering and discriminant analysis.
Abstract: Mixmod is a well-established software package for fitting a mixture model of multivariate Gaussian or multinomial probability distribution functions to a given data set with either a clustering, a density estimation or a discriminant analysis purpose. The Rmixmod S4 package provides a bridge between the C++ core library of Mixmod (mixmodLib) and the R statistical computing environment. In this article, we give an overview of the model-based clustering and classification methods, and we show how the R package Rmixmod can be used for clustering and discriminant analysis.

Journal ArticleDOI
TL;DR: The major finding of this research is that the LS-SVM with the 1v1 system is the best technique for the OA_PCA features in the epileptic EEG signal classification that outperforms all the recent reported existing methods in the literature.

Proceedings ArticleDOI
07 Jun 2015
TL;DR: A novel approach to generate a binary descriptor optimized for each image patch independently inspired by the linear discriminant embedding that simultaneously increases inter and decreases intra class distances is proposed.
Abstract: In this paper we propose a novel approach to generate a binary descriptor optimized for each image patch independently. The approach is inspired by the linear discriminant embedding that simultaneously increases inter and decreases intra class distances. A set of discriminative and uncorrelated binary tests is established from all possible tests in an offline training process. The patch adapted descriptors are then efficiently built online from a subset of tests which lead to lower intra class distances thus a more robust descriptor. A patch descriptor consists of two binary strings where one represents the results of the tests and the other indicates the subset of the patch-related robust tests that are used for calculating a masked Hamming distance. Our experiments on three different benchmarks demonstrate improvements in matching performance, and illustrate that per-patch optimization outperforms global optimization.

Journal ArticleDOI
TL;DR: The origins of chemometrics within chemical pattern recognition of the 1960s and 1970s are described in this paper, with a few approaches such as PLS-DA and SIMCA becoming dominant.

Journal ArticleDOI
TL;DR: In this article, the authors used multivariate analysis of cluster analysis (CA), principal component analysis (PCA), and discriminant analysis (DA) to classify the water quality of the Muda River basin (Malaysia).
Abstract: The classification of river water quality is a useful way of reporting the water quality status of a river to control water pollution in monitored regions. The main objective of this study is to classify the water quality of the Muda River basin (Malaysia) using nine monitoring stations. This study utilised multivariate analysis of cluster analysis (CA), principal component analysis (PCA) and discriminant analysis (DA). CA and PCA identified two different clusters (classes) that reflect the different water quality characteristics of the water systems. DA validated these clusters and produced a discriminant function (DF) that can predict the cluster membership of new samples. The classification generated by the multivariate analysis is consistent with those made by the Department of Environment (DOE). This study demonstrated that multivariate statistical techniques are effective for river water classification.

Journal ArticleDOI
TL;DR: A set of R functions for the convenient handling of morphometric analysis is provided, which include data import from Excel or tab-delimited text files, descriptive statistics for populations and taxa, cluster analysis, principal component analysis, linear discriminant analysis with permutation tests, classificatory discriminantAnalysis and k-nearest neighbour classification.
Abstract: A set of R functions for the convenient handling of morphometric analysis is provided. No previous knowledge of R is required. The functions include data import from Excel or tab-delimited text files, descriptive statistics for populations and taxa, histograms of characters, correlation matrices of characters, cluster analysis, principal component analysis, linear discriminant analysis with permutation tests, classificatory discriminant analysis and k-nearest neighbour classification. The use of the functions is demonstrated on a sample data set. Detailed descriptions of the functions and examples of the scripts for producing graphics are included as an electronic appendix. Documentation and function definitions can be downloaded from http://www.prf.jcu.cz/systematics/morphotools.html .

Journal ArticleDOI
TL;DR: The approach not only extracts sufficient spectral-spatial features from original hyperspectral images but also gets better feature representation owing to tensor framework.
Abstract: We propose to integrate spectral–spatial feature extraction and tensor discriminant analysis for hyperspectral image classification. First, we apply remarkable spectral–spatial feature extraction approaches in the hyperspectral cube to extract a feature tensor for each pixel. Then, based on class label information, local tensor discriminant analysis is used to remove redundant information for subsequent classification procedure. The approach not only extracts sufficient spectral–spatial features from original hyperspectral images but also gets better feature representation owing to tensor framework. Comparative results on two benchmarks demonstrate the effectiveness of our method.