Nonlinear Feature Extraction Research Articles

Integration of RNA molecules data with prior-knowledge driven Joint Deep Semi-Negative Matrix Factorization for heart failure study.

Heart failure (HF) is the main manifestation of cardiovascular disease. Recent studies have shown that various RNA molecules and their complex connections play an essential role in HF’s pathogenesis and pathological progression. This paper aims to mine key RNA molecules associated with HF. We proposed a Prior-knowledge Driven Joint Deep Semi-Negative Matrix Factorization (PD-JDSNMF) model that uses a hierarchical nonlinear feature extraction method that integrates three types of data: mRNA, lncRNA, and miRNA. The PPI information is added to the model as prior knowledge, and the Laplacian constraint is used to help the model resist the noise in the genetic data. We used the PD-JDSNMF algorithm to identify significant co-expression modules. The elements in the module are then subjected to bioinformatics analysis and algorithm performance analysis. The results show that the PD-JDSNMF algorithm can robustly select biomarkers associated with HF. Finally, we built a heart failure diagnostic model based on multiple classifiers and using the Top 13 genes in the significant module, the AUC of the internal test set was up to 0.8714, and the AUC of the external validation set was up to 0.8329, which further confirmed the effectiveness of the PD-JDSNMF algorithm.

HyperNTF: A hypergraph regularized nonnegative tensor factorization for dimensionality reduction

Tensor decomposition is an effective tool for learning multi-way structures and heterogeneous features from high-dimensional data, such as the multi-view images and multichannel electroencephalography (EEG) signals, are often represented by tensors. However, most of tensor decomposition methods are the linear feature extraction techniques, which are unable to reveal the nonlinear structure within high-dimensional data. To address such problem, a lot of algorithms have been proposed for simultaneously performs linear and non-linear feature extraction. A representative algorithm is the Graph Regularized Nonnegative Matrix Factorization (GNMF) for image clustering. However, the normal 2-order graph can only model the pairwise similarity of objects, which cannot sufficiently exploit the complex structures of samples. Thus, we propose a novel method, named Hypergraph Regularized Nonnegative Tensor Factorization (HyperNTF), which utilizes hypergraph to model the complex connections among samples and employs the factor matrix corresponding with last mode of Canonical Polyadic (CP) decomposition as low-dimensional representation of original data. Extensive experiments on synthetic manifolds, real-world image datasets, and EEG signals, demonstrating that HyperNTF outperforms the state-of-the-art methods in terms of dimensionality reduction, clustering, and classification.

Deep Morphological Neural Networks

Mathematical morphology intends to extract object features such as geometric and topological structures in digital images. Given a set of target images and original images, it is cumbersome and time-consuming to determine the suitable morphological operations and structuring elements. In this paper, we propose deep morphological neural networks, which include a nonlinear feature extraction layer to learn the structuring element correctly and an adaptive layer to select appropriate morphological operations automatically. We demonstrate the applications of object recognition, including hand-written digits, geometric shapes, traffic signs, and brain tumor. Experimental results show the higher computational efficiency and higher accuracy of our developed model as compared against existing convolutional neural network models.

Deep learning of complex process data for fault classification based on sparse probabilistic dynamic network

BackgroundThe dynamic and nonlinear characteristics of process data have become the major issue in data-driven process monitoring. Traditional data-driven methods are often only able to extract a single feature in process data. Therefore, how to effectively extract multi-dimensional features has become the focus of current research. MethodsSparse probabilistic dynamic network (SPDN) is a deep learning model proposed in this paper for the purpose of fault classification. The method mainly takes the advantages of the sparse Gaussian-Bernoulli Restricted Boltzmann Machine (GRBM) and the recurrent neural network (RNN) with long-short term memory (LSTM) units. First, the sparse GRBM is used for nonlinear feature extraction in an unsupervised way. Then, LSTM is introduced to realize the modeling of sequence data which can effectively handle the dynamic feature of the data. FindingsIn the Tennessee-Eastman benchmark process, the classification accuracies of the proposed method are proved to be far superior to MLP, RNN and PDN. Meanwhile, in order to prove the influence of the data dynamics and the internal parameters of the structure on the fault classification results, two additional experiments were carried out.

Risk control of metal raw materials based on deep learning

Stabilizing the purchase cost of metal raw materials is of great significance to the metal manufacturing industry. Most enterprises use futures hedging strategies to cope with the risks arising from fluctuations in the prices of metal raw materials. However, the difference between spot and futures prices of metals makes it impossible to fully control the risk. In order to further improve the efficiency of hedging and controlling corporate risks, it is necessary to accurately predict futures prices. However, the decomposition algorithms in traditional mixed models are prone to modal aliasing and have limited ability to extract nonlinear features from futures prices. Therefore, this paper proposes a variational modal decomposition-sample entropy-Cascaded Long Short-Term Memory Neural Network Model (VMD-SE-CLSTM). This paper proposes SE combined with VMD algorithm to determine the decomposition number to suppress the aliasing phenomenon of subsequence patterns, and introduces CLSTM network to improve the extraction ability of nonlinear features in futures data. The experiments are compared with 10 mainstream model methods and the method proposed in this paper. The experimental results show that the model reduces the prediction error and improves the prediction accuracy of the model, which is of great significance for enterprises to improve hedging efficiency, reduce operating risks and control production costs.

An automated solid waste detection using the optimized YOLO model for riverine management.

Due to urbanization, solid waste pollution is an increasing concern for rivers, possibly threatening human health, ecological integrity, and ecosystem services. Riverine management in urban landscapes requires best management practices since the river is a vital component in urban ecological civilization, and it is very imperative to synchronize the connection between urban development and river protection. Thus, the implementation of proper and innovative measures is vital to control garbage pollution in the rivers. A robot that cleans the waste autonomously can be a good solution to manage river pollution efficiently. Identifying and obtaining precise positions of garbage are the most crucial parts of the visual system for a cleaning robot. Computer vision has paved a way for computers to understand and interpret the surrounding objects. The development of an accurate computer vision system is a vital step toward a robotic platform since this is the front-end observation system before consequent manipulation and grasping systems. The scope of this work is to acquire visual information about floating garbage on the river, which is vital in building a robotic platform for river cleaning robots. In this paper, an automated detection system based on the improved You Only Look Once (YOLO) model is developed to detect floating garbage under various conditions, such as fluctuating illumination, complex background, and occlusion. The proposed object detection model has been shown to promote rapid convergence which improves the training time duration. In addition, the proposed object detection model has been shown to improve detection accuracy by strengthening the non-linear feature extraction process. The results showed that the proposed model achieved a mean average precision (mAP) value of 89%. Hence, the proposed model is considered feasible for identifying five classes of garbage, such as plastic bottles, aluminum cans, plastic bags, styrofoam, and plastic containers.

Condition-driven probabilistic adversarial autoencoder with nonlinear Gaussian feature learning for nonstationary process monitoring

For industrial processes, the observed data usually has strong nonlinearity and non-Gaussian properties. The whole industrial process shows typical nonstationary characteristics due to frequent time-varying behaviors. This makes the nonstationary process monitoring a challenging task along time direction. It is recognized that the nonstationary process characteristic can change with a certain rule along condition direction instead of time direction. In the present work, a novel condition-driven probabilistic adversarial autoencoder (CPAAE) algorithm is designed to address this problem. CPAAE divides the whole nonstationary process into multiple condition slices by cutting the indicator variables of the industrial process into multiple equal intervals. Subsequently, probabilistic adversarial autoencoder (PAAE) models for different condition slices can be established to extract nonlinear Gaussian features. The condition slices will be aggregated into different condition modes by evaluating the similarity of Gaussian features, and multiple monitoring models can be established for different condition modes to replace the conventional time-driven method. In this way, the nonstationary changes along the time direction can be restored to different condition modes, revealing similar process characteristics in the same condition mode. Finally, the nonstationary industrial process can be monitored by checking the changes of both Gaussian features and reconstruction errors for different condition modes. A numerical case and a real thermal power plant process are adopted to validate the feasibility of the proposed method.

Magnetic Tile Surface Defect Detection Methodology Based on Self-Attention and Self-Supervised Learning.

As the core component of permanent magnet motor, the magnetic tile defects seriously affect the quality of industrial motor. Automatic recognition of the surface defects of the magnetic tile is a difficult job since the patterns of the defects are complex and diverse. The existing defect recognition methods result in difficulty in practical application due to the complicated system structure and the low accuracy of the image segmentation and the target detection for the diversity of the defect patterns. A self-supervised learning (SSL) method, which benefits from its nonlinear feature extraction performance, is proposed in this study to improve the existing approaches. We proposed an efficient multihead self-attention method, which can automatically locate single or multiple defect areas of magnetic tile and extract features of the magnetic tile defects. We also designed an accurate full-connection classifier, which can accurately classify different defects of magnetic tile defects. A knowledge distillation process without labeling is proposed, which simplifies the self-supervised training process. The process of our method is as follows. A feature extraction model consists of standard vision transformer (ViT) backbone, which is trained by contrast learning without labeled dataset that is used to extract global and local features from the input magnetic tile images. Then, we use a full-connection neural network, which is trained by using labeled dataset to classify the known defect types. Finally, we combined the feature extraction model and defect classification model together to form a relatively simple integrated system. The public magnetic tile surface defect dataset, which holds 5 defect categories and 1 nondefect category, is used in the process of training, validating, and testing. We also use online data augmentation techs to increase training samples to make the model converge and achieve high classification accuracy. The experimental results show that the features extracted by the SSL method can get richer and more detailed features than the supervised learning model gets. The composite model reaches to a high testing accuracy of 98.3%, and gains relatively strong robustness and good generalization ability.

SlimTrain---A Stochastic Approximation Method for Training Separable Deep Neural Networks

Deep neural networks (DNNs) have shown their success as high-dimensional function approximators in many applications; however, training DNNs can be challenging in general. DNN training is commonly phrased as a stochastic optimization problem whose challenges include nonconvexity, nonsmoothness, insufficient regularization, and complicated data distributions. Hence, the performance of DNNs on a given task depends crucially on tuning hyperparameters, especially learning rates and regularization parameters. In the absence of theoretical guidelines or prior experience on similar tasks, this requires solving a series of repeated training problems which can be time-consuming and demanding on computational resources. This can limit the applicability of DNNs to problems with nonstandard, complex, and scarce datasets, e.g., those arising in many scientific applications. To remedy the challenges of DNN training, we propose \tt slimTrain, a stochastic optimization method for training DNNs with reduced sensitivity to the choice of hyperparameters and fast initial convergence. The central idea of slimTrain is to exploit the separability inherent in many DNN architectures; that is, we separate the DNN into a nonlinear feature extractor followed by a linear model. This separability allows us to leverage recent advances made for solving large-scale, linear, ill-posed inverse problems. Crucially, for the linear weights, \tt slimTrain does not require a learning rate and automatically adapts the regularization parameter. In our numerical experiments using function approximation tasks arising in surrogate modeling and dimensionality reduction, slimTrain outperforms existing DNN training methods with the recommended hyperparameter settings and reduces the sensitivity of DNN training to the remaining hyperparameters. Since our method operates on mini-batches, its computational overhead per iteration is modest and savings can be realized by reducing the number of iterations (due to quicker initial convergence) or the number of training problems that need to be solved to identify effective hyperparameters.

A DLSTM-Network-Based Approach for Mechanical Remaining Useful Life Prediction.

Remaining useful life prediction is one of the essential processes for machine system prognostics and health management. Although there are many new approaches based on deep learning for remaining useful life prediction emerging in recent years, these methods still have the following weaknesses: (1) The correlation between the information collected by each sensor and the remaining useful life of the machinery is not sufficiently considered. (2) The accuracy of deep learning algorithms for remaining useful life prediction is low due to the high noise, over-dimensionality, and non-linear signals generated during the operation of complex systems. To overcome the above weaknesses, a general deep long short memory network-based approach for mechanical remaining useful life prediction is proposed in this paper. Firstly, a two-step maximum information coefficient method was built to calculate the correlation between the sensor data and the remaining useful life. Secondly, the kernel principal component analysis with a simple moving average method was designed to eliminate noise, reduce dimensionality, and extract nonlinear features. Finally, a deep long short memory network-based deep learning method is presented to predict remaining useful life. The efficiency of the proposed method for remaining useful life prediction of a nonlinear degradation process is demonstrated by a test case of NASA’s commercial modular aero-propulsion system simulation data. The experimental results also show that the proposed method has better prediction accuracy than other state-of-the-art methods.

Kernel principal component analysis and differential non-linear feature extraction of pesticide residues on fruit surface based on surface-enhanced Raman spectroscopy.

Surface-enhanced Raman spectroscopy (SERS) has attracted much attention because of its high sensitivity, high speed, and simple sample processing, and has great potential for application in the field of pesticide residue detection. However, SERS is susceptible to the influence of a complex detection environment in the detection of pesticide residues on the surface of fruits, facing problems such as interference from the spectral peaks of detected impurities, unclear dimension of effective correlation data, and poor linearity of sensing signals. In this work, the enhanced raw data of the pesticide thiram residues on the fruit surface using gold nanoparticle (Au-NPs) solution are formed into the raw data set of Raman signal in the IoT environment of Raman spectroscopy principal component detection. Considering the non-linear characteristics of sensing data, this work adopts kernel principal component analysis (KPCA) including radial basis function (RBF) to extract the main features for the spectra in the ranges of 653∼683 cm−1, 705∼728 cm−1, and 847∼872 cm−1, and discusses the effects of different kernel function widths (σ) to construct a qualitative analysis of pesticide residues based on SERS spectral data model, so that the SERS spectral data produce more useful dimensionality reduction with minimal loss, higher mean squared error for cross-validation in non-linear scenarios, and effectively weaken the interference features of detecting impurity spectral peaks, unclear dimensionality of effective correlation data, and poor linearity of sensing signals, reflecting better extraction effects than conventional principal component analysis (PCA) models.

A multi-scale and multi-domain heart sound feature-based machine learning model for ACC/AHA heart failure stage classification

Objective. Heart sounds can reflect detrimental changes in cardiac mechanical activity that are common pathological characteristics of chronic heart failure (CHF). The ACC/AHA heart failure (HF) stage classification is essential for clinical decision-making and the management of CHF. Herein, a machine learning model that makes use of multi-scale and multi-domain heart sound features was proposed to provide an objective aid for ACC/AHA HF stage classification. Approach. A dataset containing phonocardiogram (PCG) signals from 275 subjects was obtained from two medical institutions and used in this study. Complementary ensemble empirical mode decomposition and tunable-Q wavelet transform were used to construct self-adaptive sub-sequences and multi-level sub-band signals for PCG signals. Time-domain, frequency-domain and nonlinear feature extraction were then applied to the original PCG signal, heart sound sub-sequences and sub-band signals to construct multi-scale and multi-domain heart sound features. The features selected via the least absolute shrinkage and selection operator were fed into a machine learning classifier for ACC/AHA HF stage classification. Finally, mainstream machine learning classifiers, including least-squares support vector machine (LS-SVM), deep belief network (DBN) and random forest (RF), were compared to determine the optimal model. Main results. The results showed that the LS-SVM, which utilized a combination of multi-scale and multi-domain features, achieved better classification performance than the DBN and RF using multi-scale or/and multi-domain features alone or together, with average sensitivity, specificity, and accuracy of 0.821, 0.955 and 0.820 on the testing set, respectively. Significance. PCG signal analysis provides efficient measurement information regarding CHF severity and is a promising noninvasive method for ACC/AHA HF stage classification.

Autoencoders reloaded

In Bourlard and Kamp (Biol Cybern 59(4):291–294, 1998), it was theoretically proven that autoencoders (AE) with single hidden layer (previously called “auto-associative multilayer perceptrons”) were, in the best case, implementing singular value decomposition (SVD) Golub and Reinsch (Linear algebra, Singular value decomposition and least squares solutions, pp 134–151. Springer, 1971), equivalent to principal component analysis (PCA) Hotelling (Educ Psychol 24(6/7):417–441, 1993); Jolliffe (Principal component analysis, springer series in statistics, 2nd edn. Springer, New York ). That is, AE are able to derive the eigenvalues that represent the amount of variance covered by each component even with the presence of the nonlinear function (sigmoid-like, or any other nonlinear functions) present on their hidden units. Today, with the renewed interest in “deep neural networks” (DNN), multiple types of (deep) AE are being investigated as an alternative to manifold learning Cayton (Univ California San Diego Tech Rep 12(1–17):1, 2005) for conducting nonlinear feature extraction or fusion, each with its own specific (expected) properties. Many of those AE are currently being developed as powerful, nonlinear encoder–decoder models, or used to generate reduced and discriminant feature sets that are more amenable to different modeling and classification tasks. In this paper, we start by recalling and further clarifying the main conclusions of Bourlard and Kamp (Biol Cybern 59(4):291–294, 1998), supporting them by extensive empirical evidences, which were not possible to be provided previously (in 1988), due to the dataset and processing limitations. Upon full understanding of the underlying mechanisms, we show that it remains hard (although feasible) to go beyond the state-of-the-art PCA/SVD techniques for auto-association. Finally, we present a brief overview on different autoencoder models that are mainly in use today and discuss their rationale, relations and application areas.

RF Fingerprint Extraction Method Based on CEEMDAN and Multidomain Joint Entropy

Specific emitter identification (SEI) can distinguish communication radio emitters with the fingerprint features carried by the received signal, and this technology has been widely used in military and civilian fields. However, in the real electromagnetic environment, the number of communication radio emitters is large and the signal-to-noise ratio (SNR) is low, which leads to poor nonlinear fingerprint analysis of SEI in a single domain. Therefore, combining the exploration of multiple domains of electromagnetic spatial information resources, this paper proposed a radio frequency (RF) fingerprint extraction method based on complementary ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and multidomain joint entropy. The proposed method is an attempt and exploration further extraction of nonlinear fingerprint features in multiple domains. Firstly, considering the nonstationarity of the communication signal, this article adopts the CEEMDAN method to decompose the signal to multiple intrinsic mode functions (IMF). Then, the decomposed signal is represented in multiple spaces by a multidimensional phase space reconstruction technique. Nonlinear analysis of the original signal is performed in multiple spaces: multidimensional differential approximate entropy space, singular spectral entropy space, and power spectral entropy space. Finally, the support vector machine (SVM) is adopted in the classification stage. To demonstrate the robustness of the method, the method is verified on the universal software radio peripheral (USRP) dataset and the Northeastern University public dataset. In terms of the identification accuracy, the proposed method performs with 98.5% accuracy on the 5-class USRP dataset. It also performs with 94.7% accuracy on the 16-class public dataset. The experimental results show that the proposed method has a stable identification performance and has a more than 85% recognition rate in the SNR above 5dB.

Cooperative multinetworks semi-supervised pre-stack seismic inversion

SUMMARY The elastic properties of the subsurface, such as density, P-velocity and S-velocity, can be estimated in pre-stack seismic inversion. In recent research, the deep neural network is widely used in pre-stack seismic inversion for its strong non-linear fitting and feature extraction ability. However, the label data is generally inadequate due to high drilling costs and strong data-sharing barriers in the field of exploration geophysics. In order to reduce the dependence of network performance on label data and ensure the accuracy of inversion mostly, semi-supervised learning is adopted. Here, we develop a cooperative multinetworks semi-supervised pre-stack seismic inversion method. In the cooperative multinetworks inversion framework, the inversion network, mapping network, and modification network are adopted to complete the inversion task cooperatively. A forward network is constructed to automatically generate seismic data from density, P-velocity and S-velocity, which can assist the above networks to complete semi-supervised learning. Compared with some published deep learning pre-stack inversion methods, the spatio-temporal correlation of data can be fully mined, the prior geological structure and low-frequency information can be utilized effectively, and reflectivity is adopted as an intermediate output parameter to improve the robustness of the method. The experiments on the Marmousi2 model demonstrate that cooperative multinetworks semi-supervised inversion strategy is superior to conventional semi-supervised inversion methods in both inversion accuracy and antinoise performance. In addition, the susceptibility experiments of the initial model indicate that the proposed method can maintain a high inversion accuracy with little effective information in the initial model. Finally, the proposed method is successfully applied to the field data and obtains a high-resolution inversion result.

Hyperspectral Anomaly Detection Using Deep Learning: A Review

Hyperspectral image-anomaly detection (HSI-AD) has become one of the research hotspots in the field of remote sensing. Because HSI’s features of integrating image and spectrum provide a considerable data basis for abnormal object detection, HSI-AD has a huge application potential in HSI analysis. It is difficult to effectively extract a large number of nonlinear features contained in HSI data using traditional machine learning methods, and deep learning has incomparable advantages in the extraction of nonlinear features. Therefore, deep learning has been widely used in HSI-AD and has shown excellent performance. This review systematically summarizes the related reference of HSI-AD based on deep learning and classifies the corresponding methods into performance comparisons. Specifically, we first introduce the characteristics of HSI-AD and the challenges faced by traditional methods and introduce the advantages of deep learning in dealing with these problems. Then, we systematically review and classify the corresponding methods of HSI-AD. Finally, the performance of the HSI-AD method based on deep learning is compared on several mainstream data sets, and the existing challenges are summarized. The main purpose of this article is to give a more comprehensive overview of the HSI-AD method to provide a reference for future research work.

Convolutional neural network long short-term memory deep learning model for sonic well log generation for brittleness evaluation

Brittleness index, usually calculated by P- and S-wave velocities, is an important parameter used to optimize the sweet spots of shale oil. Empirical relationships or artificial intelligence networks can predict sonic logs based on conventional logging data, but accuracy is limited by the formation types and properties, such as shale sandstone interbeds. Therefore, we propose a hybrid convolutional neural network long short-term memory (CNN-LSTM) deep learning model for the prediction of compressional and shear traveltimes. The new model can extract nonlinear features as well as fluctuating trends of log response features with depth, which is different from most machine learning methods that only consider extracting spatial features between logs or only time-series features of log data sets. We conclude that the new CNN-LSTM network has the highest prediction accuracy (92% and 91.3%) and advantages in predicting curve mutation points compared with other machine learning models. We apply the predicted compressional and shear sonic times for evaluating brittleness to reduce the risk of exploration in shale oil reservoirs.

Prediction of Remaining Useful Life Using Fused Deep Learning Models: A Case Study of Turbofan Engines

Abstract The study of intelligent operation and maintenance methods for turbofan engines is of great importance for improving the reliability of turbofan engines. Given the harsh operating conditions and complex structure of the turbofan engine, it is extremely difficult to establish an accurate physical model for remaining useful life (RUL) prediction. The traditional operation and maintenance method based on the physical model has several limitations in the application of turbofan engines, while the data-driven method offers a new solution. Compared with traditional machine learning models, deep learning models possess more powerful nonlinear expression capabilities and feature extraction capabilities. Therefore, this study focuses on studying the RUL prediction algorithm for turbofan engines based on the fused deep learning models. In this article, a multimodal deep learning approach based on a 1DCNN (1D convolutional neural network) + attention enhanced Bi-LSTM (bidirectional long short-term memory) network is proposed to predict the RUL by mining the temporal information of data. Furthermore, a DDResNet (dilated deep residual network) is also introduced to the 1DCNN submodel to leverage its hidden pattern mining capability due to its advance in preventing performance degradation across layers. Subsequently, the output of these two submodels is weighted fused to obtain the final RUL prediction. The merits of the proposed method are demonstrated by comparing it with existing methods for RUL prediction using the C-MAPSS (commercial modular aero-propulsion system simulation) dataset.

Aerodynamic system instability identification with sample entropy algorithm based on feature extraction

Based on the sample entropy algorithm in nonlinear dynamics, an improved sample entropy method is proposed in the aerodynamic system instability identification for the stall precursor detection based on the nonlinear feature extraction algorithm in an axial compressor. The sample entropy algorithm is an improved algorithm based on the approximate entropy algorithm, which quantifies the regularity and the predictability of data in time series. Combined with the spatial modes representing for the rotating stall in the circumferential direction, the recognition capacity of the sample entropy is displayed well on the detection of stall inception. The indications of rotating waves are extracted by the circumferential analysis from modal wave energy. The significant ascendant in the amplitude of the spatial mode is a pronounced feature well before the imminence of stall. Data processing with the spatial mode effectively avoids the problems of inaccurate identification of a single measuring point only depending on pressure. Due to the different selections of similarity tolerance, two kinds of sample entropy are obtained. The properties of the development process of the identification model show obvious mutation phenomena at the boundary of instability, which reveal the inherent characteristic in aerodynamic system. Then the dynamic difference quotient is computed according to the difference quotient criterion, after the smooth management by discrete wavelet. The rapid increase of difference quotient can be regarded as a significant feature of the system approaching the flow instability. It is proven that based on the principle of sample entropy algorithm, the nonlinear characteristic of rotating stall can be well described. The inception can be suggested by about 12–68 revolutions before the stall arrival. This prediction method presenting is accounted for the nonlinearity of the complex flow in stall, which is in a view of data fusion system of pressure for the spatial mode tracking.

Research on Face Recognition Algorithm Based on Image Processing.

While network technology is convenient for our daily life, the problems that are exposed are also endless. The most important thing for everyone is information security. In order to improve the security level of network information and identify and detect faces, the method used in this paper has improved compared with the traditional AdaBoost method and skin color method. AdaBoost detection is performed on the image, which reduces the probability of false detection. The experiment compares the experimental results of the AdaBoost method, the skin color method and the skin color + AdaBoost method. All operations in the KPCA and KFDA algorithms are performed by the inner product kernel function defined in the original space, and no specific non-linear mapping function is involved.The full name of KPCA is kernel principal component analysis. The full name of KFDA is kernel Fisher discriminant analysis. Combining the zero-space method kernel discriminant analysis method improves the ability of discriminant analysis to extract non-linear features. Through the secondary extraction of PCA features, a better recognition result than the PCA method is obtained. This paper also proposes a zero-space based Fisher discriminant analysis method. Experiments show that the zero-space-based method makes full use of the useful discriminant information in the zero space of the intraclass dispersion matrix, which improves the accuracy of face recognition to some extent.If you choose the polynomial kernel function, when d = 0.8, KPCA has a higher recognition ability. When d = 2, the recognition rate of KFDA and zero space-based KFDA is the largest. For polynomial functions, in general, d = 2.