Linear Feature Extraction Methods Research Articles

Surface and underwater acoustic target recognition using only two hydrophones based on machine learning.

Surface and underwater (S/U) acoustic targets recognition is an important application of passive sonar. It is difficult to distinguish them due to the mixture of underwater target radiation noise and marine environmental noise. In previous studies, although using a single hydrophone was able to identify S/U acoustic targets, there were still a few hydrophones that had poor accuracy. In this paper, S/U acoustic targets recognition using two hydrophones based on Gradient Boosting Decision Tree is proposed, and it is first found out as high as 100% accuracy could be achieved with the implementation of SACLANT 1993 data. The real experimental data are always rare and insufficient. The big training dataset is generated using environmental information by acoustic model named KRAKEN. Simulation and experimental data used in the model are heterogeneous, and the differences between these two kinds of data are assimilated by using vertical linear array feature extraction method. The model realizes the recognition of S/U acoustic targets based on channel information besides source spectrum information. By using the combination of two hydrophones, the surface and underwater targets recognition accuracy reached 1 and 0.9384, while they are only 0.4715 and 0.5620 using a single hydrophone, respectively.

Systematic comparison of unsupervised feature extraction methods for enhancing discrimination accuracy of brown rice flour adulteration in LIBS

Laser-induced breakdown spectroscopy (LIBS) is widely applied in various critical fields due to its technical advantages. However, with the explosive growth in spectral data volume, the challenges of high-dimension and redundant information of spectra are becoming increasingly prominent. Therefore, dimensionality reduction of spectra is crucial for LIBS. In this study, to enhance the discrimination accuracy of brown rice flour and its adulterants, a systematic comparative analysis of eight unsupervised feature extraction methods was conducted, including principal component analysis (PCA), non-negative matrix factorization (NMF), independent component analysis (ICA), multidimensional scaling (MDS), kernel PCA (KPCA), locally linear embedding (LLE), Laplacian eigenmaps (LE), and isometric mapping (ISOMAP). The results indicate that linear feature extraction methods (PCA, NMF, ICA, and MDS) show a more obvious improvement in the analytical performance of the classification models, compared to nonlinear feature extraction methods (KPCA, LLE, LE, and ISOMAP). Among these methods, the commonly used PCA has the best improvement effect. The accuracy, macro precision, macro recall, and macro F1-score of the classification models improve from 92.73%, 94.13%, 92.92%, and 0.93 to 95.52%, 95.67%, 95.63%, and 0.96. Therefore, this study can provide experience and guidance for selecting appropriate unsupervised feature extraction methods in subsequent research, contributing to the further application and development of LIBS.

Alzheimer's disease classification based on nonlinear high-order features and hypergraph convolutional neural network

Alzheimer's disease (AD) is an irreversible neurodegenerative disorder that damages patients' memory and cognitive abilities. Therefore, the diagnosis of AD holds significant importance. The interactions between regions of interest (ROIs) in the brain often involve multiple areas collaborating in a nonlinear manner. Leveraging these nonlinear higher-order interaction features to their fullest potential contributes to enhancing the accuracy of AD diagnosis. To address this, a framework combining nonlinear higher-order feature extraction and three-dimensional (3D) hypergraph neural networks is proposed for computer-assisted diagnosis of AD. First, a support vector machine regression model based on the radial basis function kernel was trained on ROI data to obtain a base estimator. Then, a recursive feature elimination algorithm based on the base estimator was applied to extract nonlinear higher-order features from functional magnetic resonance imaging (fMRI) data. These features were subsequently constructed into a hypergraph, leveraging the complex interactions captured in the data. Finally, a four-dimensional (4D) spatiotemporal hypergraph convolutional neural network model was constructed based on the fMRI data for classification. Experimental results on the Alzheimer's Disease Neuroimaging Initiative (ADNI) database demonstrated that the proposed framework outperformed the Hyper Graph Convolutional Network (HyperGCN) framework by 8% and traditional two-dimensional (2D) linear feature extraction methods by 12% in the AD/normal control (NC) classification task. In conclusion, this framework demonstrates an improvement in AD classification compared to mainstream deep learning methods, providing valuable evidence for computer-assisted diagnosis of AD.

Sludge bulking monitoring in industrial wastewater treatment plants through graphical methods: A dynamic graph embedding and Bayesian networks approach

Sludge bulking is a prevalent issue in wastewater treatment plants (WWTPs) that negatively impacts effluent quality by hindering the normal functioning of treatment processes. To tackle this problem, we propose a novel graph-based monitoring framework that employs advanced graph-based techniques to detect and diagnose sludge bulking events. The proposed framework utilizes historical datasets under normal operating conditions to extract pertinent features and causal relationships between process variables. This enables operators to trigger alarms and diagnose the root cause of the bulking event. Sludge bulking detection is carried out using the dynamic graph embedding (DGE) method, which identifies similarities among process variables in both temporal and neighborhood dependencies. Consequently, the dynamic Bayesian network (DBN) computes the prior and posterior probabilities of a belief, updated at each time step. Variations in these probabilities indicate the potential root cause of the sludge bulking event. The results demonstrate that the DGE outperforms other linear and non-linear feature extraction methods, achieving a detection rate of 99%, zero false alarms, and less than one percent incorrect detections. Additionally, the DBN-based diagnostic method accurately identified the majority of sludge bulking root causes, primarily those resulting from sudden drops in COD concentration, with an accuracy of 98% an improvement of 11% over state-of-the-art techniques.

Learning latent local manifold embeddings for hyperspectral feature extraction

Hyperspectral images with an immense number of spectral bands provide abundant discriminant information for accurate land-cover classification in the remote sensing field. However, these narrow and adjacent bands contain a large amount of redundant information. Analyzing these images always requires a huge storage space with expensive computational costs. Furthermore, their high correlation coefficient would lead to the Hughes phenomenon, hindering the improvement of classification performance. We propose a linear semi-supervised hyperspectral feature extraction method L3ME to learn latent local manifold embeddings. Although labeled samples are beneficial to construct learning models, their number is always limited in real-world tasks. The motivation of this paper is to jointly enhance the contributions of labeled and unlabeled samples for learning local manifold structures of hyperspectral images. Features in labeled samples are extracted by two procedures, the adaptive patch alignment framework and integrated intraclass-interclass relationships, from different perspectives. The former aims to solve the problem of the uneven distribution of classes by introducing spectral angle based adaptive parameters. The latter aims to solve the problem of the uneven distribution of samples by constructing several adjacency graphs. The locality preserving projection is capable of preserving the local neighborhood structure of samples. A penalty for sparse regularization is cleverly integrated into the proposed linear discriminant objective function, which is optimized using a novel updating strategy. The convergence of L3ME is proved in detail and analyzed in this paper. Experiments on three typical hyperspectral datasets illustrate the effectiveness of the proposed method over some state-of-the-art techniques. The implementation of L3ME is available at https://github.com/biowby/L3ME.

Natural gas pipeline leak diagnosis based on improved variational modal decomposition and locally linear embedding feature extraction method

Natural gas pipeline leaks can cause serious hazards to natural gas transportation and pose considerable risks to the environment and the safety of residents. Therefore, feature extraction of pipeline signals is crucial in natural gas pipeline leak detection. However, the quality of feature extraction directly affects the effectiveness of pipeline leak detection. Therefore, this paper proposes a pipeline leakage feature extraction method based on variable mode decomposition (VMD) and local linear embedding (LLE). First, the pipeline signal is decomposed into several modal components by VMD; then, the dispersion entropy is used to select the feature modes. Secondly, the time-frequency domain features of different components are extracted to construct a high-dimensional feature matrix, which LLE reduces to obtain the classified low-dimensional feature vectors. Finally, the extracted feature vectors are used to train and test the support vector machine (SVM). By analyzing the experimental results, it can be seen that the proposed method can classify pipeline signals with an accuracy of up to 95%, which effectively solves the problem of false alarms and missed alarms in pipeline leakage detection.

Application of Coupled LDA–KPCA and BO–MKRVM Model to Predict Coal and Gas Outbursts

This paper proposed a coupled model of effective feature extraction and optimized classifier, which can overcome the existing problems of coal and gas outbursts classification in the literatures. Firstly, we support the use of kernel principal component analysis and linear discriminant analysis methods to extract linear and nonlinear feature information from coal and gas outbursts influencing factors. Secondly, in order to realize the complementarity and correlation between coal and gas outbursts influencing factors, we perform parallel feature fusion to combine the extracted linear and nonlinear feature information. Thirdly, an improved classifier called BO–MKRVM has been proposed that combines mixed kernel relevance vector machine (MKRVM) and Bayesian optimization (BO) algorithm to predict coal and gas outbursts according to the extracted features. In BO–MKRVM model, an effective mixed kernel function which combines Gaussian kernel function and Sigmoid kernel function is proposed to improve the learning and generalization ability of the MKRVM classifier, then the BO and tenfold cross-validation are utilized to optimize kernel parameters and weight coefficients of MKRVM with strong global and local search capability, the proposed BO–MKRVM classifier is performed on coal and gas outbursts dataset. Compared with the single feature extraction method, the combination of linear and non-linear feature extraction methods can obtain complete feature information and contribute to the classification performance of outbursts. The mixed kernel function considers the characteristics of coal and gas outbursts sample data, which can also effectively improve the outbursts accuracy. After the MKRVM classifier is optimized by BO algorithm, the BO–MKRVM classifier has better fitting effect and generalization ability, and obtains higher accuracy with a lower time. The experimental results are obviously better than those of other classification models, which further verifies the applicability of the proposed coupled model.

A comparison of Laryngeal effect in the dialects of Punjabi language

Human beings have their own speaking style which helped them in depicting their native language. The major reason behind variability in some language is due to varying dialect of the speakers. In the field of Automatic Speech Recognition (ASR), key challenge is to recognize and to generate an acoustic model which represents differences of redundant acoustic features. In this paper, an issue of dialect classification is perform on the basis of tonal aspects of laryngeal phoneme [h]. This is an empirical study of [h] sound words in four major dialects of Indian Punjabi language with two key parameters, namely F0 variation, and acoustic space, which are calculated using two formant frequencies: F1, and F2. The results are based on four different dialects which provide us some interesting hypotheses and are explored with self-created dataset. The speech analysis tool PRAAT features have been extracted and correlations are studied using Statistical Package for the Social Sciences (SPSS). Each variable has been compared with same variable of all other dialects. The results analysis showed that the fundamental frequency of these vowels are influenced distinctly in different dialectal conditions. Apart F1 and F2 have shown a significant correlation with each spoken dialect. Further work is extended through processing of acoustic information at feature level or by comparing the performance analysis using basic or hybrid Linear Predictive Cepstral Coefficients feature extraction methods. The result shows that the hybrid LPCC + F0 system achieved a Relative Improvement (R.I.) of 6.94% on Subspace Gaussian Mixture Model model in comparison to that of basic LPCC approach respectively.

Analysis of the Information Transfer Rate-ITR in Linear and Non-linear Feature Extraction Methods for SSVEP Signals

The most popular paradigm in BCIs is the steady-state visually evoked potential (SSVEP) due to their advantages, such as the high information transfer rate (ITR), the time spent on users in the training phase, and the capacity to discriminate each stimulus. One of the most influential factors in the ITR evaluation is the feature extraction methods since these can increase the accuracy. Here, we compare nine methods for the extraction of features from SSVEP signals to identify those with better performance, according to the time window (TW), its technology (equipment and number of nodes), and the value of ITR. The study identifies two groups: the first one is characterized by presenting variations of correlated component analysis (CCA), which is highly used to increase the ITR due to its efficiency in classification and its capacity of response to reduction (TW), such as MsetCCA, IT-CCA, FBCCA; the second one are the representation special based methods that consider the non-linear nature of the electroencephalogram (EEG) signal such as TRCA, CORRCA, EMD, and VMD. The results show a considerable difference between these groups. The maximum ITR value for FBCCA was 117.75 [bits/min] in a TW of 1.25s, while the VMD method achieved 3120 [bits/min] in a TW of 1s, respectively. The comparison covers signals between 0.55 and 8 seconds, taking into account visual strain, the experimental environment, and other artifacts.

Linear and exponential fault-assistant feature extraction methods for process monitoring

Feature extraction is very important for dimension reduction and facilitation of the final monitoring accuracy. In practice, both the normal data and fault data can be easily collected and stored by the advanced sensing and computer technologies. The measured fault data can also be employed to improve monitoring performance, which was not yet fully researched in the past. This work proposes novel fault-assistant monitoring models of both linear and exponential versions. In these two versions of models, the measured normal and fault data are used together to extract the common features, and then special features can be further extracted based on the normal residuals. Then the process data can be divided into common, special, and residual subspaces. The monitoring statistics can be correspondingly defined for fault detection. The numerical case and two industrial cases are applied to show the efficiency and feasibility of the proposed method.

Information-theoretic feature selection with segmentation-based folded principal component analysis (PCA) for hyperspectral image classification

ABSTRACT Hyperspectral image (HSI) usually holds information of land cover classes as a set of many contiguous narrow spectral wavelength bands. For its efficient thematic mapping or classification, band (feature) reduction strategies through Feature Extraction (FE) and/or Feature Selection (FS) methods for finding the intrinsic bands’ information are typically applied. Principal Component Analysis (PCA) is a frequently employed unsupervised linear FE method whereas cumulative-variance accumulation is used for selecting top features from PCA data. However, PCA can fail to extract intrinsic structure of HSI due to global variance consideration and domination by visible and near infrared bands while cumulative-variance accumulation has no capability to exploit non-linear relationships among the transformed features produced by PCA-based FE methods. Consequently, we propose an information theoretic normalized Mutual Information (nMI)-based minimum Redundancy Maximum Relevance (mRMR) non-linear measure to select the intrinsic features from the transformed space of our previously proposed Segmented-Folded-PCA (Seg-Fol-PCA) and Spectrally Segmented-Folded-PCA (SSeg-Fol-PCA) FE methods. We extensively analyse the effectiveness of the proposed unsupervised FE and supervised FS combinations Seg-Fol-PCA-mRMR and SSeg-Fol-PCA-mRMR with that of PCA-based existing linear and non-linear state-of-the-art methods. In addition, cumulative variance-based top features pick-up strategy is considered with all FE methods and Renyi quadratic entropy-based FS is used with Kernel Entropy Component Analysis (Ker-ECA). The experimental results illustrate that SSeg-Fol-PCA-mRMR and Seg-Fol-PCA-mRMR obtain highest classification result e.g. 95.39% and 95.03% respectively for agricultural Indian Pines HSI, and 96.58% and 95.30% respectively for urban Washington DC Mall HSI while the classification accuracies using all original features of the HSIs are 70.28% and 91.90% respectively. Moreover, the proposed SSeg-Fol-PCA-mRMR and Seg-Fol-PCA-mRMR outperform all investigated combinations of FE and FS using the real HSI datasets.

Efficient deep discriminant embedding: Application to face beauty prediction and classification

Inspired by deep learning architectures, we introduce a multi-layer local discriminant embedding algorithm that integrates feature selection as a main step to capture the most relevant and discriminant features of an input face image or face descriptor. The proposed framework allows to transform any linear method to a deep variant via a cascaded feature extraction and selection architecture able to convert weak and noisy descriptors to strong ones. As a case study, the local discriminant embedding (LDE) projection is adopted as a linear feature extraction method. The resulting framework can be considered as an efficient deep discriminant embedding technique. To validate this framework, we have considered two different computer vision problems: face beauty prediction which involves both classification and regression tasks, and face recognition which is a classical classification problem. Experiments conducted on different public benchmark databases show that this approach enhances the performance of the LDE algorithm and provides a discriminating strategy to solve the dimensionality reduction problem. For face beauty regression, our proposed framework achieved on average an improvement of about 5% and 7% with respect to two other configurations where only VGG-face and VGG-face followed by LDE have been considered. For face beauty classification, the proposed algorithm outperformed many classical manifold learning techniques reaching in some databases improvements of about 10%.

Writer Identification of Lampung Handwritten Documents Based on Selected Characters

Writer identification is a sub-field in handwriting recognition which its objective is to determine the identity of the writer based on handwriting input. The goal is usually for forensic purposes such as finding the perpetrators of crimes that leave traces of evidence in the form of written messages. In addition, writer identification can also be used to determine the identity of a historical actor if he or she leaves a valuable written artefact. The object of this research is the traditional character of the Lampung region which is so-called Had Lampung by the local community. The traditional character of Lampung consists of 20 main characters and 12 diacritics. Based on selected characters, the writer will be recognized using the Principal Component Analysis (PCA) feature. PCA is one linear feature extraction method of an object in pattern recognition. The PCA algorithm consists of several stages, namely the calculation of the average dataset, the subtraction of the vector dataset with averages, the calculation of covariance, the calculation of eigenvectors and eigenvalues, eigenvector reduction, and the projection of the dataset against reduced eigenvector space. PCA in this paper is used as a feature in image recognition. The dataset utilized in this study is the Lampung Dataset which is a handwritten character recognition (HWCR) dataset. Lampung Dataset consists of 82 Lampung handwritten documents. All Lampung character images in the dataset were extracted from these documents using the connected component extraction algorithm and eventually generated 32,140 images. Furthermore, these images are converted into grayscale images. In this research, as many as 12,500 grayscale images of Lampung handwriting characters were chosen to represent 82 different writers. This data is employed as training and testing data on the proposed method. The highest accuracy of the identification of the writer using this PCA feature is 82.92%, while the lowest accuracy is 28.29%.

Entropy: A Promising EEG Biomarker Dichotomizing Subjects With Opioid Use Disorder and Healthy Controls.

Electroencephalography (EEG) signals are known to be nonstationary and often multicomponential signals containing information about the condition of the brain. Since the EEG signal has complex, nonlinear, nonstationary, and highly random behaviour, numerous linear feature extraction methods related to the short-time windowing technique do not satisfy higher classification accuracy. Since biosignals are highly subjective, the symptoms may appear at random in the time scale and very small variations in EEG signals may depict a definite type of brain abnormality it is valuable and vital to extract and analyze the EEG signal parameters using computers. The challenge is to design and develop signal processing algorithms that extract this subtle information and use it for diagnosis, monitoring, and treatment of subjects suffering from psychiatric disorders. For this purpose, finite impulse response-based filtering process was employed rather than traditional time and frequency domain methods. Finite impulse response subbands were analyzed further to obtain feature vectors of different entropy markers and these features were fed into a classifier namely multilayer perceptron. The performances of the classifiers were finally compared considering overall classification accuracies, area under receiver operating characteristic curve scores. Our results underline the potential benefit of the introduced methodology is promising and is to be treated as a clinical interface in dichotomizing substance use disorders subjects and for other medical data analysis studies. The results also indicate that entropy estimators can distinguish normal and opioid use disorder subjects. EEG data and theta frequency band have distinctive capability for almost all types of entropies while nonextensive Tsallis entropy outperforms compared with other types of entropies.

Kernel PCA for road traffic data non‐linear feature extraction

Road traffic data bring great challenges for data processing and traffic-state analysis. The feature extraction is an effective way to make full use of road traffic data. Here, the authors explore the temporal correlation of road traffic data in time series and a new non-linear feature extraction method for road traffic data, namely, kernel principal component analysis (PCA). The low-dimensional road traffic data are mapped into the high-dimensional space based on a kernel function, and then the PCA method is applied to conduct feature extraction in the high-dimensional space. To have a clear comparison, the authors compare the kernel PCA method with a linear PCA-based feature extraction method. The results show that the extracted features under kernel PCA are more useful than PCA and the kernel PCA method has a better compression effect than PCA under the same condition.

Fault Detection Method of Luojia1-01 Satellite Attitude Control System Based on Supervised Local Linear Embedding

This paper is aimed at the telemetry data of Luojia1-01 satellite attitude control system, and carries out the research on high-dimensional data feature extraction and fault detection based on supervised local linear embedding (SLLE). Because the general linear feature extraction method can not mine the feature information of nonlinear high-dimensional telemetry data, a data feature extraction and fault detection method based on local linear embedding is designed. Combined with the statistics SPE and T 2 , the low-dimensional feature information is obtained for data statistics and monitoring faults. Due to the local linear embedded manifold learning method for the traditional batch processing mode is difficult to update and improve the database online, and the supervised local linear embedding method is introduced. The online sample feature extraction and fault detection schemes are designed, and the database is updated by updating the weight matrix through online samples. Finally, the effectiveness of the method is verified by Luojia1-01 satellite telemetry data. The results show that the fault detection method based on SLLE reduces the false alarm rate (FAR) by approximately 3% and the missing alarm rate (MAR) by approximately 10% compared with local linear embedding (LLE). This method effectively improves the detection capability of the anomalous state of Luojia1-01 satellite and has certain engineering application value.

Classification of EEG Signals Using Hybrid Feature Extraction and Ensemble Extreme Learning Machine

Electroencephalogram (EEG) signals play an important role in clinical diagnosis and cognitive neuroscience. Automatic classification of EEG signals is gradually becoming the research focus, which contains two procedures: feature extraction and classification. In the phase of feature extraction, a hybrid feature extraction method is proposed and the features are derived by performing linear and nonlinear feature extraction methods, which can describe abundant properties of original EEG signals. In order to eliminate irrelevant and redundant features, feature selection based on class separability is employed to select the optimal feature subset. In the phase of classification, this paper presents a novel ensemble extreme learning machine based on linear discriminant analysis. Linear discriminant analysis is used to transform training subsets that are generated by bootstrap method, through which we can increase the differences of basic classifiers and reduce generalization errors of ensemble extreme learning machine. Experiments on two different EEG datasets are conducted in this study. Class separability is investigated to verify the effectiveness of feature extraction methods. The overall classification results show that compared with other similar studies, the proposed method can significantly enhance the performance of EEG signals classification.

Non-destructive identification of maize haploid seeds using nonlinear analysis method based on their near-infrared spectra

The identification of maize haploid seeds is a significant process in genetic research and modern maize breeding. Adopting near-infrared spectroscopy technology to distinguish haploid seeds from hybrid seeds has the advantages of being non-destructive, rapid and low cost. However, due to the influence of light, temperature, humidity, near-infrared intensity, instrument and dynamic change of seed activity, the near-infrared spectra of maize seeds showed high dimensional nonlinear characteristics. In this study, to make full use of the class label information, a nonlinear feature analytical method for haploid maize seeds identification based on Supervised Virtual Sample Kernel Locality Preserving Projection (SVSKLPP) has been proposed. The experimental results showed that the nonlinear identification model SVSKLPP achieved strong classification performance to identify the maize haploid seeds. Moreover, compared with the linear feature extraction methods Principal Component Analysis, Orthogonal Linear Discriminant Analysis, Locality Preserving Projection, Supervised Virtual Sample Locality Preserving Projection and nonlinear feature extraction methods Kernel Locality Preserving Projection, Isomap, Locally Linear Embedding, Laplacian Eigenmaps and Local Tangent Space Alignment, the SVSKLPP model achieved a better performance. The average accuracy, sensitivity and specificity using method SVSKLPP were 97.1%, 98.8% and 95.4% respectively, and it also had high robustness. The overall results show that the SVSKLPP-NIR methodology was efficient in accurately identifying haploid maize seeds, thus demonstrating its capabilities for application in haploid breeding for crop variety improvement.

IntensityPatches and RegionPatches for image recognition

In this paper, we propose a framework for defining feature extraction techniques, called Pixel Clustering. It is an extension of feature extraction with Wavelets. We propose two linear feature extraction techniques using Pixel Clustering: IntensityPatches and RegionPatches. We assess the methods in color and grayscale image datasets: two face datasets and two object datasets. The proposed methods present a short computation time for feature extraction and high accuracy compared with linear feature extraction methods and other state-of-the-art feature extraction techniques.

Hyperspectral anomaly detection based on stacked denoising autoencoders

Hyperspectral anomaly detection (AD) is an important technique of unsupervised target detection and has significance in real situations. Due to the high dimensionality of hyperspectral data, AD will be influenced by noise, nonlinear correlation of band, or other factors that lead to the decline of detection accuracy. To overcome this problem, a method of hyperspectral AD based on stacked denoising autoencoders (AE) (HADSDA) is proposed. Simultaneously, two different feature detection models, spectral feature (SF) and fused feature by clustering (FFC), are constructed to verify the effectiveness of the proposed algorithm. The SF detection model uses the SF of each pixel. The FFC detection model uses a similar set of pixels constructed by clustering and then fuses the set of pixels by the stacked denoising autoencoders algorithm (SDA). The SDA is an algorithm that can automatically learn nonlinear deep features of the image. Compared with other linear or nonlinear feature extraction methods, the detection result of the proposed algorithm is greatly improved. Experiment results show that the proposed algorithm is an excellent feature learning method and can achieve higher detection performance.