Extract Low-dimensional Features Research Articles

Remaining useful life prediction for rolling bearings using multi-layer grid search and LSTM

Estimating the remaining useful life (RUL) of equipment is critical for ensuring the safe operation of machinery and reducing maintenance losses. For the existing RUL prediction, the problem of data redundancy and initial prediction time dramatically affects the prediction results. Therefore, this paper proposes a long short-term memory network (LSTM) RUL prediction algorithm that is based on multi-layer grid search (MLGS) optimization. This method integrates feature data and optimizes network parameters to ensure accuracy and effectively predict the non-stationary degradation of the bearing. Firstly, this paper uses a data fusion method to extract low dimensional feature vectors of running data, and the multi-feature fusion is performed to obtain the principal component index. Then, the MLGS is used to optimize the network overparameters. The effect of the initial measurement time on prediction results can be reduced, and the calculation speed can be improved. Finally, the IMS data set of NASA is used for verification and comparative testing. The test results show that the proposed RUL prediction algorithm can effectively reduce the influence of the initial prediction time on the prediction accuracy compared with other prediction methods.

Hybrid deep-learning framework for object-based forgery detection in video

Detection of object-based video forgery is receiving attention in recent years. However, until recently, dominant object-based forgery detectors are still based on hand-crafted features and their performances are not satisfactory. In this paper, we propose a novel hybrid deep-learning network which for the first time incorporates two-dimensional/three-dimensional convolutional neural network and recurrent neural network for object-based video forgery detection in videos with advanced encoding formats. Please note that the proposed framework is a full end-to-end data-driven solution. It is comprised of a specifically initialized three-dimensional convolutional layer which tries to mix up primitive intra-frame and inter-frame features, a two-dimensional CNN which tries to extract low-dimensional intra-frame features, and a four-layer bi-directional LSTM network which tries to catch high-level temporal features. Using this way, our proposed approach catches the intra-frame and inter-frame inherent properties of a target video with a united framework. The extensive experiments conducted on the largest object-based forged video database ever reported in the literature show that our hybrid framework achieves superior performance in forged video detection and forged segment localization. Moreover, the experiments conducted on datasets of videos with degraded quality demonstrated that our proposed framework is more robust in real-life scenarios.

A Lightweight System Towards Viewing Angle and Clothing Variation in Gait Recognition

In this paper, the authors have proposed a computationally efficient, robust, and lightweight system for gait recognition. The proposed system contains two main stages: In the first stage, a classification network identifies optical flow corners in the normalized silhouette and calculates the distances traveled in every viewpoint which is further used by a regression model to identify the viewing angle. In the second stage, a feature extraction network computes the gait energy image (GEI) for every viewpoint and then uses principal component analysis (PCA) to extract low dimensional feature vectors from these GEI images. Finally, a multi-layer perceptron model is trained using the extracted principal components for every viewing angle. The performance of a system is comprehensively evaluated on the CASIA B and OULP gait dataset. The experimental results demonstrate the superior performance of a proposed system in viewing angle classification (100% accuracy), gait recognition (100% accuracy in normal walk), computational efficiency, robustness to clothing, and viewing angle variation.

A Novel Band Selection and Spatial Noise Reduction Method for Hyperspectral Image Classification

As an essential reprocessing method, dimensionality reduction (DR) can reduce the data redundancy and improve the performance of hyperspectral image (HSI) classification. A novel unsupervised DR framework with feature interpretability, which integrates both band selection (BS) and spatial noise reduction method, is proposed to extract low-dimensional spectral-spatial features of HSI. We proposed a new Neighboring band Grouping and Normalized Matching Filter (NGNMF) for BS, which can reduce the data dimension whilst preserve the corresponding spectral information. An enhanced 2-D singular spectrum analysis (E2DSSA) method is also proposed to extract the spatial context and structural information from each selected band, aiming to decrease the intra-class variability and reduce the effect of noise in the spatial domain. The support vector machine (SVM) classifier is used to evaluate the effectiveness of the extracted spectral-spatial low-dimensional features. Experimental results on three publicly available HSI datasets have fully demonstrated the efficacy of the proposed NGNMF-E2DSSA method, which has surpassed a number of state-of-the-art DR methods.

Quantized Tensor Neural Network

Tensor network as an effective computing framework for efficient processing and analysis of high-dimensional data has been successfully applied in many fields. However, the performance of traditional tensor networks still cannot match the strong fitting ability of neural networks, so some data processing algorithms based on tensor networks cannot achieve the same excellent performance as deep learning models. To further improve the learning ability of tensor network, we propose a quantized tensor neural network in this article (QTNN), which integrates the advantages of neural networks and tensor networks, namely, the powerful learning ability of neural networks and the simplicity of tensor networks. The QTNN model can be further regarded as a generalized multilayer nonlinear tensor network, which can efficiently extract low-dimensional features of the data while maintaining the original structure information. In addition, to more effectively represent the local information of data, we introduce multiple convolution layers in QTNN to extract the local features. We also develop a high-order back-propagation algorithm for training the parameters of QTNN. We conducted classification experiments on multiple representative datasets to further evaluate the performance of proposed models, and the experimental results show that QTNN is simpler and more efficient while compared to the classic deep learning models.

Uncovering spatial representations from spatiotemporal patterns of rodent hippocampal field potentials

SummarySpatiotemporal patterns of large-scale spiking and field potentials of the rodent hippocampus encode spatial representations during maze runs, immobility, and sleep. Here, we show that multisite hippocampal field potential amplitude at ultra-high-frequency band (FPAuhf), a generalized form of multiunit activity, provides not only a fast and reliable reconstruction of the rodent's position when awake, but also a readout of replay content during sharp-wave ripples. This FPAuhf feature may serve as a robust real-time decoding strategy from large-scale recordings in closed-loop experiments. Furthermore, we develop unsupervised learning approaches to extract low-dimensional spatiotemporal FPAuhf features during run and ripple periods and to infer latent dynamical structures from lower-rank FPAuhf features. We also develop an optical flow-based method to identify propagating spatiotemporal LFP patterns from multisite array recordings, which can be used as a decoding application. Finally, we develop a prospective decoding strategy to predict an animal's future decision in goal-directed navigation.

Using one-class autoencoder for adulteration detection of milk powder by infrared spectrum

Food adulteration detection requires quick and simple methods. Spectral detection can significantly reduce the analysis time, but it needs to construct a detection model. In this study, a one-class classification method based on an autoencoder is proposed for the detection of food adulteration by spectroscopy. In the proposed method, the autoencoder is constructed to extract low-dimensional features from high-dimensional spectral data and reconstruct the original spectrum. Then the coding error and reconstruction error are used to determine the food sample is adulterated or not. The infrared spectral data of milk powder and its adulterated forms are used to verify the performance of the proposed model. Experimental results show that the proposed method has similar effects to soft independent modeling of class analogy and one-class partial least squares, and is significantly better than support vector data description. The proposed method can be flexibly applied to the spectral detection of food adulteration.

TNNL: A novel image dimensionality reduction method for face image recognition

Face image recognition technology plays an important role in biometric recognition field. Among the present face recognition methods, the method based on subspace learning have aroused wide concern due to their favorable properties, such as convenience for computation and effectiveness for identification. However, existing methods based on subspace learning are out of work when the sample-specific corruptions and outliers come along. To solve this problem, we build a novel model for face image recognition named truncated nuclear norm on low rank discriminant embedding (TNNL). The TNNL can mitigate the negative impact of noise and enhance the discriminability of features. Furthermore, we propose two iterative algorithms to extract the robust low dimensional image feature. To testify the effectiveness and robustness of TNNL, we conduct experiments on two benchmark face image databases for low dimensional feature extraction. The experimental results show that TNNL is better than the existing methods.

Predicting circRNA-disease associations based on autoencoder and graph embedding

Circular RNAs (circRNAs) are a special kind of non-coding RNA. They play important regulatory role in diseases through interactions of miRNAs associated with the diseases. Due to their insensitivity to nucleases, they are more stable than linear RNAs. It is thus imperative to integrate available information for predicting circRNA-disease associations in humans. Here, we propose a computational model to predict circRNA-disease associations based on accelerated attributed network embedding (AANE) algorithm and autoencoder(AE). First, we use AANE algorithm to extract low-dimensional features of circRNAs and diseases and then stacked autoencoder (SAE) to automatically extract in-depth features. The features obtained by AANE and the SAE are integrated and XGBoost is used as a binary classifier to get the predicted results. The proposed model has an average area under the receiver operating characteristic curve value of 0.8800 in 5-fold cross validation and 0.8988 in 10-fold cross validation. The factors that can affect the performance of the model are discussed and some common diseases are used as case studies. Results indicated that the model has great performance in predicting circRNA-disease associations.

A cost-effective manufacturing process recognition approach based on deep transfer learning for CPS enabled shop-floor

The rapid development of the industrial Internet of Things has promoted manufacturing to develop towards the cyber-physical system, of which highly accurate process recognition plays an important role in achieving proactive monitoring of intelligent manufacturing process. Compared to the traditional handcrafted feature-based method, deep model owns convenience in terms of extracting feature automatically for the recognition. However, training a deep model is time-consuming and also requires large-scale training samples. To solve these problems and obtain high accuracy in the meanwhile, a deep transfer learning-based manufacturing process recognition approach is proposed in this study. A pre-trained model based on a convolutional neural network is used to extract low dimensional features followed by a fine-tuning process to target the specific process recognition task. Experimental verification of two datasets was conducted to demonstrate this cost-effective method. The results showed the proposed method can get better accuracy with less training time and fewer training samples.

Low rank representation and discriminant analysis-based models for peer-to-peer default risk assessment

PurposeThis study aims to assess the default risk of borrowers in peer-to-peer (P2P) online lending platforms. The authors propose a novel default risk classification model based on data cleaning and feature extraction, which increases risk assessment accuracy.Design/methodology/approachThe authors use borrower data from the Lending Club and propose the risk assessment model based on low-rank representation (LRR) and discriminant analysis. Firstly, the authors use three LRR models to clean the high-dimensional borrower data by removing outliers and noise, and then the authors adopt a discriminant analysis algorithm to reduce the dimension of the cleaned data. In the dimension-reduced feature space, machine learning classifiers including the k-nearest neighbour, support vector machine and artificial neural network are used to assess and classify default risks.FindingsThe results reveal significant noise and redundancy in the borrower data. LRR models can effectively clean such data, particularly the two LRR models with local manifold regularisation. In addition, the supervised discriminant analysis model, termed the local Fisher discriminant analysis model, can extract low-dimensional and discriminative features, which further increases the accuracy of the final risk assessment models.Originality/valueThe originality of this study is that it proposes a novel default risk assessment model, based on data cleaning and feature extraction, for P2P online lending platforms. The proposed approach is innovative and efficient in the P2P online lending field.

Selecting XFEL single-particle snapshots by geometric machine learning.

A promising new route for structural biology is single-particle imaging with an X-ray Free-Electron Laser (XFEL). This method has the advantage that the samples do not require crystallization and can be examined at room temperature. However, high-resolution structures can only be obtained from a sufficiently large number of diffraction patterns of individual molecules, so-called single particles. Here, we present a method that allows for efficient identification of single particles in very large XFEL datasets, operates at low signal levels, and is tolerant to background. This method uses supervised Geometric Machine Learning (GML) to extract low-dimensional feature vectors from a training dataset, fuse test datasets into the feature space of training datasets, and separate the data into binary distributions of “single particles” and “non-single particles.” As a proof of principle, we tested simulated and experimental datasets of the Coliphage PR772 virus. We created a training dataset and classified three types of test datasets: First, a noise-free simulated test dataset, which gave near perfect separation. Second, simulated test datasets that were modified to reflect different levels of photon counts and background noise. These modified datasets were used to quantify the predictive limits of our approach. Third, an experimental dataset collected at the Stanford Linear Accelerator Center. The single-particle identification for this experimental dataset was compared with previously published results and it was found that GML covers a wide photon-count range, outperforming other single-particle identification methods. Moreover, a major advantage of GML is its ability to retrieve single particles in the presence of structural variability.

Assessment of unsteady flow predictions using hybrid deep learning based reduced-order models

In this paper, we present two deep learning-based hybrid data-driven reduced-order models for prediction of unsteady fluid flows. These hybrid models rely on recurrent neural networks (RNNs) to evolve low-dimensional states of unsteady fluid flow. The first model projects the high-fidelity time series data from a finite element Navier–Stokes solver to a low-dimensional subspace via proper orthogonal decomposition (POD). The time-dependent coefficients in the POD subspace are propagated by the recurrent net (closed-loop encoder–decoder updates) and mapped to a high-dimensional state via the mean flow field and the POD basis vectors. This model is referred to as POD-RNN. The second model, referred to as the convolution recurrent autoencoder network (CRAN), employs convolutional neural networks (instead of POD) as layers of linear kernels with nonlinear activations, to extract low-dimensional features from flow field snapshots. The flattened features are advanced using a recurrent (closed-loop manner) net and up-sampled (transpose convoluted) gradually to high-dimensional snapshots. Two benchmark problems of the flow past a cylinder and the flow past side-by-side cylinders are selected as the unsteady flow problems to assess the efficacy of these models. For the problem of the flow past a single cylinder, the performance of both the models is satisfactory and the CRAN model is found to be overkill. However, the CRAN model completely outperforms the POD-RNN model for a more complicated problem of the flow past side-by-side cylinders involving the complex effects of vortex-to-vortex and gap flow interactions. Owing to the scalability of the CRAN model, we introduce an observer-corrector method for calculation of integrated pressure force coefficients on the fluid–solid boundary on a reference grid. This reference grid, typically a structured and uniform grid, is used to interpolate scattered high-dimensional field data as snapshot images. These input images are convenient in training the CRAN model, which motivates us to further explore the application of the CRAN-based models for prediction of fluid flows.

Real-time Process Authentication for Additive Manufacturing Processes based on In-situ Video Analysis

Additive manufacturing (AM) processes are subject to cyber-physical attacks during all the three stages including design, slicing, and manufacturing phases. In-situ process authentication is crucial for AM to ensure that the manufacturing is performed as intended. Since most of the cyber-physical attacks aiming to alter AM processes can be manifested in the change of printing path, an in-situ optical imaging system is capable of detecting alteration in printing path in real time through texture analysis. This will prevent catastrophic geometric changes and mechanical property compromises in the AM parts, and hence improving the AM process security. In this paper, a new part authentication framework is proposed by leveraging layer-wise in-situ videos. The distribution of the segmented textures’ geometric features is extracted from the layer-wise videos, and the multilinear principal component analysis (MPCA) algorithm is used to extract low-dimensional features from the joint distribution of geometric features. A case study based on a fused filament fabrication (FFF) process is used to validate the proposed framework.

A novel feature learning framework for high-dimensional data classification

Feature extraction is an essential component in many classification tasks. Popular feature extraction approaches especially deep learning-based methods, need large training samples to achieve satisfactory performance. Although dictionary learning-based methods are successfully used for feature extraction on both small and large datasets, however, when dealing with high-dimensional datasets, a large number of dimensions also mask the discriminative information embedded in the data. To address these issues, a novel feature learning framework for high-dimensional data classification is proposed in this paper. Specially, to discard the irrelevant parts that derail the dictionary learning process, the dictionary is adaptively learnt in the low-dimensional space parameterized by a transformation matrix. To ensure that the learned features are discriminative for the classifier, the classification results in turn are used to guide the dictionary and transformation matrix learning process. Compared with other methods, the proposed method simultaneously exploits the dimension reduction, dictionary learning and classifier learning in one optimization framework, which enables the method to extract low-dimensional and discriminative features. Experimental results on several benchmark datasets demonstrate the superior performance of the proposed method for high-dimensional data classification task, particularly when the number of training samples is small.

Speech Separation Using Convolutional Neural Network and Attention Mechanism

Speech information is the most important means of human communication, and it is crucial to separate the target voice from the mixed sound signals. This paper proposes a speech separation model based on convolutional neural networks and attention mechanism. The magnitude spectrum of the mixed speech signals, as the input, has its high dimensionality. By analyzing the characteristics of the convolutional neural network and attention mechanism, it can be found that the convolutional neural network can effectively extract low-dimensional features and mine the spatiotemporal structure information in the speech signals, and the attention mechanism can reduce the loss of sequence information. The accuracy of speech separation can be improved effectively by combining two mechanisms. Compared to the typical speech separation model DRNN-2 + discrim, this method achieves 0.27 dB GNSDR gain and 0.51 dB GSIR gain, which illustrates that the speech separation model proposed in this paper has achieved an ideal separation effect.

Deep density-based image clustering

Recently, deep clustering, which is able to perform feature learning that favors clustering tasks via deep neural networks, has achieved remarkable performance in image clustering applications. However, the existing deep clustering algorithms generally need the number of clusters in advance, which is usually unknown in real-world tasks. In addition, the initial cluster centers in the learned feature space are generated by k-means. This only works well on spherical clusters and probably leads to unstable clustering results. In this paper, we propose a two-stage deep density-based image clustering (DDC) framework to address these issues. The first stage is to train a deep convolutional autoencoder (CAE) to extract low-dimensional feature representations from high-dimensional image data, and then apply t-SNE to further reduce the data to a 2-dimensional space favoring density-based clustering algorithms. In the second stage, we propose a novel density-based clustering technique for the 2-dimensional embedded data to automatically recognize an appropriate number of clusters with arbitrary shapes. Concretely, a number of local clusters are generated to capture the local structures of clusters, and then are merged via their density relationship to form the final clustering result. Experiments demonstrate that the proposed DDC achieves comparable or even better clustering performance than state-of-the-art deep clustering methods, even though the number of clusters is not given.

A Deep Regression Model with Low-Dimensional Feature Extraction for Multi-Parameter Manufacturing Quality Prediction

Manufacturing quality prediction can be used to design better parameters at an earlier production stage. However, in complex manufacturing processes, prediction performance is affected by multi-parameter inputs. To address this issue, a deep regression framework based on manifold learning (MDRN) is proposed in this paper. The multi-parameter inputs (i.e., high-dimensional information) were firstly analyzed using manifold learning (ML), which is an effective nonlinear technique for low-dimensional feature extraction that can enhance the representation of multi-parameter inputs and reduce calculation burdens. The features obtained through the ML were then learned by a deep learning architecture (DL). It can learn sufficient features of the pattern between manufacturing quality and the low-dimensional information in an unsupervised framework, which has been proven to be effective in many fields. Finally, the learned features were inputted into the regression network, and manufacturing quality predictions were made. One type (two cases) of machinery parts manufacturing system was investigated in order to estimate the performance of the proposed MDRN with three comparisons. The experiments showed that the MDRN overwhelmed all the peer methods in terms of mean absolute percentage error, root-mean-square error, and threshold statistics. Based on these results, we conclude that integrating the ML technique for dimension reduction and the DL technique for feature extraction can improve multi-parameter manufacturing quality predictions.

Research on feature selection for rotating machinery based on Supervision Kernel Entropy Component Analysis with Whale Optimization Algorithm

Aimed at finding a scientific and effective method to diagnose faults in rotating machinery, the algorithm based on Supervision Kernel Entropy Component Analysis with Whale Optimization Algorithm (WOSKECA) for feature selection has been proposed in this paper. Firstly, for ensure sufficient information gathering from the rotating machinery, multi-features parameters of the time domain, frequency domain, time-frequency domain and entropy domain are extracted, and a high-dimensional feature matrix is constructed from these features. Secondly, the WOSKECA for feature selection is applied to eliminate any redundant information. The algorithm takes the class information as the supervised information to improve the recognition accuracy of Kernel Entropy Component Analysis (KECA) and can extract low-dimensional features with discriminative ability from the high-dimensional feature space. Meanwhile, the Whale Optimization Algorithm (WOA) as a new meta-heuristic optimization algorithm is applied to optimize the kernel parameters in KECA, which reduces the interference of subjective factors and reduces the professionalism of obtaining fault information. Finally, Support Vector Machine based on the Particle Swarm Optimization (PSOSVM) is used to classify the fault type as well as assess the severity of the faults. The feature extraction algorithm is entirely evaluated through experimentation and comparative. The results show that the proposed method is able to detect and classify the faults of rotating machinery more successfully and more accurately than traditional manifold learning.

Heart sound classification using the SNMFNet classifier

Objective: Heart sound classification still suffers from the challenges involved in achieving high accuracy in the case of small samples. Dimension reduction attempts to extract low-dimensional features with more discriminability from high-dimensional spaces or raw data, and is popular in learning predictive models that target small sample problems. However, it can also be harmful to classification, because any reduction has the potential to lose information containing category attributes. Approach: For this, a novel SNMFNet classifier is designed to directly associate the dimension reduction process with the classification procedure used for promoting feature dimension reduction to follow the approach that is beneficial for classification, thus making the low-dimensional features more distinguishable and addressing the challenge facing heart sound classification in small samples. Main results: We evaluated our method and representative methods using a public heart sound dataset. The experimental results demonstrate that our method outperforms all comparative models with an obvious improvement in small samples. Furthermore, even if used with relatively sufficient samples, our method performs at least as well as the baseline that uses the same high-dimensional features. Significance: The proposed SNMFNet classifier significantly to improves the small sample problem in heart sound classification.