Wavelet Kernel Research Articles

Blockchain and artificial intelligence enabled privacy‐preserving medical data transmission in Internet of Things

AbstractAdvancements in information technology have benefited the healthcare industry by providing it with distinct methods of managing medical data which improve the quality of medical services. The Internet of Things (IoT) and artificial intelligence are the foundations for innovative sustainable computing technologies in e‐healthcare applications. In the IoT‐enabled sustainable healthcare system, the IoT devices normally record the patient data and transfer it to the cloud for further processing. Security is considered an important issue in the design of IoT networks in the healthcare environment. To resolve this issue, this article presents a novel blockchain and artificial intelligence‐enabled secure medical data transmission (BAISMDT) for IoT networks. The goal of the BAISMDT model is to achieve security and privacy in reliable data transmission of the IoT networks. The proposed model involves a signcryption technique for secure and reliable IoT data transmission. The blockchain‐enabled secure medical data transmission process takes place among the IoT gadgets and service providers. The blockchain technique is applied to generate a viable environment to securely and reliably transmit data among different data providers. Next to the decryption process, the modified discrete particle swarm optimization algorithm with wavelet kernel extreme learning machine model is applied to determine the presence of disease. An extensive set of simulations were carried out on a benchmark medical dataset. The experimental results analysis pointed out the superior performance of the proposed BAISMDT model with the accuracy of 97.54% and 98.13% on the applied Heart Statlog and WBC dataset, respectively.

An Effective Turkey Marble Classification System: Convolutional Neural Network with Genetic Algorithm -Wavelet Kernel - Extreme Learning Machine

Marble is one of the most popular decorative elements. Marble quality varies depending on its vein patterns and color, which are the two most important factors affecting marble quality and class. The manual classification of marbles is likely to lead to various mistakes due to different optical illusions. However, computer vision minimizes these mistakes thanks to artificial intelligence and machine learning. The present study proposes the Convolutional Neural Network- (CNN-) with genetic algorithm- (GA) Wavelet Kernel- (WK-) Extreme Learning Machine (ELM) (CNN–GA-WK-ELM) approach. Using CNN architectures such as AlexNet, VGG-19, SqueezeNet, and ResNet-50, the proposed approach obtained 4 different feature vectors from 10 different marble images. Later, Genetic Algorithm (GA) was used to optimize adjustable parameters, i.e. k, 1, and m, and hidden layer neuron number in Wavelet Kernel (WK) – Extreme Learning Machine (ELM) and to increase the performance of ELM. Finally, 4 different feature vector parameters were optimized and classified using the WK-ELM classifier. The proposed CNN–GA-WK-ELM yielded an accuracy rate of 98.20%, 96.40%, 96.20%, and 95.60% using AlexNet, SequeezeNet, VGG-19, and ResNet-50, respectively.

A new intelligent fault diagnosis and prognosis method for wind turbine doubly-fed induction generator

Condition Monitoring and fault-prognosis approaches are typical methods to reduce the energy production cost and Wind Turbine downtime. In this paper, a new CM combinatory system and fault prognosis are proposed based on an adaptive threshold, feature-level fusion, and new degradation indicator and the CM operation is based on a new index Symptom of Degeneration crossing of an adaptive threshold. Also, a new adaptive threshold is proposed based on the fuzzy rules and WT operation point. Fault prognosis is conducted with the Least-Squares Support-Vector Machine method, and Particle Swarm Optimization is employed for the optimum selecting of the wavelet Kernel function and the SVM parameters. The proposed technique is compared with other methods and the simulation results illustrate the PSO-LS-SVM superiorities. The effectiveness of the proposed prognostic structure is evaluated using a WT test-rig prototype. The experimental results demonstrate that the Condition-Based Maintenance is improved by the proposed structure and the RUL is predicted before serious damage occurrences.

Cooperative classification method for multi-fault diagnosis of machinery based on parameterized wavelet kernel extreme learning and sparse representation

Owing to complicated manual design and feature extraction processes of mechanical multi-fault classification and recognition methods, single-class classifiers do not adapt well to the high dimensionality and high self-similarity among mechanical vibration signals with nonlinear characteristics, resulting in insufficient classification accuracy. Combining the adaptive learning and feature extraction capabilities of the wavelet kernel-based extreme learning machine (WKELM) with the sparse representation classifier (SRC), this paper proposes a hybrid artificial bee colony (ABC)-optimized Morlet wavelet kernel-based extreme learning machine (ABC-MWKELM) with SRC. The two-level classifier is capable of cooperatively classifying multiple mechanical faults. First, a Morlet wavelet kernel-based extreme learning machine (MWKELM) model that matches the kernel function construction condition is presented. An ABC optimization algorithm is adopted to optimize critical parameters of the MWKELM model, thereby improving the accuracy and generalization ability of the model. In addition, the ABC algorithm is used to optimize the MWKELM model as a pre-classifier for classifying test samples. According to the posterior probability of each training sample category output, the threshold decision method selects candidate categories with a high confidence coefficient. When the preset meets the threshold, the classification result of the ABC-optimized MWKELM is directly adopted. Otherwise, the candidate training samples are used to construct a dictionary and the SRC is used to classify the test samples further. The experimental results demonstrate that compared with the ELM, WKELM, (ELM-SRC), and radial basis function (RBF)-KELM and SRC (RBFKELM-SRC), the method presented in this paper has significant advantages in terms of accuracy of mechanical multi-fault classification. The validity and applicability of this method in mechanical multi-fault classification were verified.

Separation of Commercially Important Tuna from Other Fishes Using Feature Descriptor and Pre-trained CNN Models

Aims / Objectives: Identification of fish species is essential in export industries. Among the different fish species exported, tuna forms a significant portion and hence the separation of tuna from other fishes is necessary. The work aims to develop automated systems for the separation of commercially important tuna from other fishes. 
 Methodology: The work proposes two models for the classification of commercial fishes. The first model uses conventional feature descriptors, which extract features from both spatial and frequency domain. These features are combined and are reduced by an ensemble dimension reduction method. The combined and reduced feature sets are evaluated using different classifiers. The second proposed model uses four pre-trained convolutional neural networks, VGG16, VGG19, Xception, and MobileNet, for the classification. The models are fine-tuned for the classification process.
 Results: Results show that for the first model, extreme learning machine classifier with Mercer wavelet kernel gives high accuracy on combined feature set while the polynomial kernel ELM provides better performance with the reduced set. For the second model, a comparison of the performance of four CNN models is done, and results indicate that VGG19 outperforms other networks in the classification task. 
 Conclusion: Among the two proposed models, pre-trained CNN based model shows better performance than the conventional method in the separation task. Different performance measures, accuracy, precision, recall, F-score, and misclassification error are used to evaluate the system. A comparison of performance of the proposed models with the state-of-the-art systems is also reported.

Application of modified wavelet and fractal modeling for detection of geochemical anomaly

Aim of this research is to propose a wavelet-fractal model for detection of geochemical anomaly based on a modified Morlet Wavelet Kernel (MMWK). This study was carried out for detection of lithium anomalies regarding stream sediment samples. Major Li anomalies commence from 51.1 ppm obtained by the wavelet-fractal modeling based on Decomposed Wavelet Transformation (DWT) using a conventional Morlet wavelet. A modified kernel for the Morlet wavelet was utilized for separation of the Li anomalies. The MMWK benefits from the use of L1 norm (instead of L2) which consequently provides more robustness and resistance to outliers for stream sediment data. Results derived via the modified Morlet kernel-fractal modeling show that there are principal lithium anomalies with Li ≥ 55.2 ppm. In addition, staged factor analysis was used and Li was placed with Rb, Cs and F in F2-2. Then, the F2-2 main anomalies commence from 2.7 and 2.9 using the conventional Morlet and MMWK, respectively. Subsequently, both methods were compared by a log-ratio matrix with an Overall Accuracy (OA) due to rock samples retaken from these anomalies. These rock samples were classified using a Concentration-Number (C-N) fractal model and high intensive anomalous samples with Li ≥ 126 ppm. This correlation indicated that the OA between the main obtained anomalies by the MMWK and the rock samples are higher than conventional Morlet-fractal modeling.

Non-destructive acoustic screening of pineapple ripeness by unsupervised machine learning and Wavelet Kernel methods.

In a pineapple exporting factory, manual lines are usually built to screen fruits of non-ripen hitting sounds from millions of undecided fruits for long-haul transportation. However, human workers cannot concentratedly listen and make consistent judgments over long hours. Pineapple screening becomes arbitrary after approximately an hour. We developed a non-destructive screening device aside from the conveyor sorter to classify pineapples automatically. The device makes intelligent judgments by tapping a sound source to the skin of pineapples and analyzing the penetrated sounds by wavelet kernel decomposition and unsupervised machine learning (ML). The sound tapping relies on the well-touch of the skin. We also design several acoustic couplers to adapt the vibrator to the skin and pick high-quality penetrated sounds. A Taguchi experiment design was used to determine the most suitable coupler. We found that our unsupervised ML method achieves 98.56% accuracy and 0.93 F1-score by using a specially designed thorn-board for assisting tapping sound to fruit skin.

Automatic identification of epileptic seizure signal using optimized added kernel support vector machine (OAKSVM)

In this work, empirical mode decomposition (EMD)-based optimized added kernel least square support vector machine (OAKLSSVM) hybridized model is proposed for automatic identification of epileptic electroencephalogram (EEG) signals where the kernel parameters are being optimized using water cycle algorithm (WCA). The proposed model with EMD decomposition and WCA optimization together is known as EMD-OAKLSSVM-WCA. Here, two kernel functions i.e., radial basis function and wavelet kernel functions are deployed together to form the added kernel framework. From EMD, intrinsic mode functions (IMFs) are obtained where Hilbert transform (HT) is used to obtain analytic form of IMFs. For classifying seizure and non-seizure EEG signals, the frequency modulation bandwidth and amplitude modulation bandwidth parameters are obtained from the analytical IMFs and are used as features for the OAKLSSVM model. The experimental results validate the efficiency of the proposed model which provides better classification accuracy (99.33%) as compared to different promising classifiers and some state-of-the-art models.

An Optimal Feature Selection with Wavelet Kernel Extreme Learning Machine for Big Data Analysis of Product Reviews

In recent times, generation of big data takes place in an exponential way from diverse textual data sources like review sites, media, blogs, etc. Sentiment analysis (SA) finds it useful to classify the opinions of the big data to different kinds ofsentiments. Therefore, SA on big data helps a business to take beneficial commercial understandings from text based content. Though several SA approaches have been presented, yet, there is a need to improve the performance of SA to interpret the customer’s feedback and increase the product quality.This paper introduces a novel social spider optimization based feature selection based wavelet kernel extreme learning machine (SSO-WKELM) model. The proposed model initially undergoes pre-processing to remove the unwanted word removal. Then, Term Frequency-Inverse Document Frequency (TF-IDF) is utilized as a feature extraction technique to extract the set of feature vectors. Besides, a social spider optimization (SSO) algorithm is utilized for feature selection process and thereby achieves improved classification performance. Subsequently, WKELM is employed as a classifier to classify the incidence of positive or negative user reviews. For experimental validation, a Product review dataset derived from Amazon along with synthetic data is used. The experimental results stated the superior classification performance of the SSO-WKELM model.

A Hydraulic Pump Fault Diagnosis Method Based on the Modified Ensemble Empirical Mode Decomposition and Wavelet Kernel Extreme Learning Machine Methods.

To address the problem that the faults in axial piston pumps are complex and difficult to effectively diagnose, an integrated hydraulic pump fault diagnosis method based on the modified ensemble empirical mode decomposition (MEEMD), autoregressive (AR) spectrum energy, and wavelet kernel extreme learning machine (WKELM) methods is presented in this paper. First, the non-linear and non-stationary hydraulic pump vibration signals are decomposed into several intrinsic mode function (IMF) components by the MEEMD method. Next, AR spectrum analysis is performed for each IMF component, in order to extract the AR spectrum energy of each component as fault characteristics. Then, a hydraulic pump fault diagnosis model based on WKELM is built, in order to extract the features and diagnose faults of hydraulic pump vibration signals, for which the recognition accuracy reached 100%. Finally, the fault diagnosis effect of the hydraulic pump fault diagnosis method proposed in this paper is compared with BP neural network, support vector machine (SVM), and extreme learning machine (ELM) methods. The hydraulic pump fault diagnosis method presented in this paper can diagnose faults of single slipper wear, single slipper loosing and center spring wear type with 100% accuracy, and the fault diagnosis time is only 0.002 s. The results demonstrate that the integrated hydraulic pump fault diagnosis method based on MEEMD, AR spectrum, and WKELM methods has higher fault recognition accuracy and faster speed than existing alternatives.

A band selection approach based on wavelet support vector machine ensemble model and membrane whale optimization algorithm for hyperspectral image

Hyperspectral Image (HSI) has become one of the important remote sensing sources for object interpretation by its abundant band information. Among them, band selection is considered as the main theme in HSI classification to reduce the data dimension, and it is a combinatorial optimization problem and difficult to be completely solved by previous techniques. Whale Optimization Algorithm (WOA) is a newly proposed swarm intelligence algorithm that imitates the predatory strategy of humpback whales, and membrane computing is able to decompose the band information into a series of elementary membranes that decreases the coding length. In addition, Support Vector Machine (SVM) combined with wavelet kernel is adapted to HSI datasets with high dimension and small samples, ensemble learning is an effective tool that synthesizes multiple sub-classifiers to solve the same problem and obtains accurate category label for each sample. In the paper, a band selection approach based on wavelet SVM (WSVM) ensemble model and membrane WOA (MWOA) is proposed, experimental results indicate that the proposed HSI classification technique is superior to other corresponding and newly proposed methods, achieves the optimal band subset with a fast convergence speed, and the overall classification accuracy has reached 93% for HSIs.

Homotopy-based hyper-heuristic searching approach for reciprocal feedback inversion of groundwater contamination source and aquifer parameters

Groundwater contamination source identification is critical for taking effective measures to design remediation strategies, assess contamination risks, and confirm contamination responsibilities. To resolve the “equifinality” problem resulting from simultaneous inversion of contamination source characteristics and aquifer parameters at dense non-aqueous phase liquid-contaminated sites, two reciprocal optimization frames for separately identifying the contamination sources and aquifer parameters were designed and connected. The two sets of identification results were corrected stepwise by means of a feedback correction iteration process, thereby sufficiently improving the identification accuracy. The ensemble learning machine (ESLM) incorporating Kriging, radical basis function neural network, support vector regression, and wavelet kernel extreme learning machine with swarm intelligence (SI) algorithm was embedded into the reciprocal inversion iterations to replace the multiphase flow simulation model for significantly improving the computational efficiency. To improve the optimization efficiency, a hyper-heuristic homotopy algorithm was constructed for segmentally searching the global optimum in wider areas with low dependence on initial values. Results showed that the combined application of SI-based ensemble learning machine (SI-ESLM) and hyper-heuristic homotopy algorithm effectively accomplished the simultaneous identification of contamination sources and aquifer parameters with high efficiency, while maintaining high accuracy. The SI-ESLM sufficiently approximated the outputs of the multiphase flow simulation model with increased certainty (R2=0.9977), while the mean relative error was limited to 1.5388%. Compared to traditional heuristic algorithms, this application of reciprocal inversion iterations and the hyper-heuristic homotopy algorithm significantly reduced the mean relative error of identification results from 6.51% to 1.03%.

Processes soft modeling based on stacked autoencoders and wavelet extreme learning machine for aluminum plant-wide application

Data-driven soft modeling has been extensively used for industrial processes to estimate key quality indicators which are hard to measure by some physical devices. However,the existing deep soft methods faces the challenge of training efficiency, gradient diminishing and explosion. Constructing an accurate and robust soft model is still a challenging topic from an application point of view. This paper develops an effective and efficient soft method (SAE-WELM) for processes modeling. First, a stacked autoencoder (SAE) is used to extract the deep features. Then, a top-layer extreme learning machine (ELM) is further applied to a plant-wide industrial aluminum production process. The activation function is wavelet kernel. Finally, the approximation and convergence of the proposed SAE-WELM are theoretically proved. The industrial case demonstrates that SAE-WELM captures the deep features faster than other iterative-based neural networks, and the accuracy and robustness outperform the existing state-of-the-art methods.

IoT Technology Enabled Heuristic Model With Morlet Wavelet Neural Network for Numerical Treatment of Heterogeneous Mosquito Release Ecosystem

The utmost advancements of artificial neural networks (ANNs), software-defined networks (SDNs) and internet of things (IoT) technologies find beneficial in different applications of the smart healthcare sector. Aiming at modern technology’s use in the future development of healthcare, this paper presents an advanced heuristic based on Morlet wavelet neural network for solving the mosquito release ecosystem in a heterogeneous atmosphere. The mosquito release ecosystem is dependent of six classes, eggs density, larvae density, pupae density, mosquitoes searching for hosts density, resting mosquito’s density and mosquitoes searching for ovipositional site density. An artificial neural network with the layer structure of Morlet wavelet (MWNN) kernel is presented using the global and local search optimization schemes of genetic algorithm (GA) and active-set algorithm (ASA), i.e., MWNN-GA-ASA. The accurateness, reliability and constancy of the proposed MWNN-GA-ASA is established through comparative examinations with Adams method based numerical results to solve the proposed nonlinear system with matching of order 10−06 to 10−09. The accuracy and convergence of the proposed MWNN-GA-ASA is certified using the statistical operators based on root mean square error (RMSE), Theil’s inequality coefficient (T.I.C) and mean absolute deviation (MAD) operators.

Improved ECG-Derived Respiration Using Empirical Wavelet Transform and Kernel Principal Component Analysis.

Many methods have been developed to derive respiration signals from electrocardiograms (ECGs). However, traditional methods have two main issues: (1) focusing on certain specific morphological characteristics and (2) not considering the nonlinear relationship between ECGs and respiration. In this paper, an improved ECG-derived respiration (EDR) based on empirical wavelet transform (EWT) and kernel principal component analysis (KPCA) is proposed. To tackle the first problem, EWT is introduced to decompose the ECG signal to extract the low-frequency part. To tackle the second issue, KPCA and preimaging are introduced to capture the nonlinear relationship between ECGs and respiration. The parameter selection of the radial basis function kernel in KPCA is also improved, ensuring accuracy and a reduction in computational cost. The correlation coefficient and amplitude square coherence coefficient are used as metrics to carry out quantitative and qualitative comparisons with three traditional EDR algorithms. The results show that the proposed method performs better than the traditional EDR algorithms in obtaining single-lead-EDR signals.

Wind speed and power prediction using MK-PINN

A multi-kernel pseudo inverse neural network (MKPINN) is proposed in this paper for efficient wind speed and power forecasting. The proposed model has been compared with Gaussian, wavelet, polynomial and sigmoid kernel. To get best output and learning methodology and stability pseudo inverse neural network is added, which substitutes the hidden layer with kernel function. This helps to achieve more accurate and faster response. Various case studies have been carried out from ten minutes to five hours interval in order to prove it accuracy.

Evaluation and Forecasting of Wind Energy Investment Risk along the Belt and Road Based on a Novel Hybrid Intelligent Model

The timely and effective investment risk assessment and forecasting are of great significance to ensure the investment safety and sustainable development of wind energy along the Belt and Road. In order to obtain the scientific and real-time forecasting result, this paper constructs a novel hybrid intelligent model based on improved cloud model combined with GRA-TOPSIS and MBA-WLSSVM. Firstly, the factors influencing investment risk of wind energy along the Belt and Road are identified from three dimensions: endogenous risk, exogenous risk and process risk. Through the fuzzy threshold method, the final input index system is selected. Secondly, the risk evaluation method based on improved cloud model and GRA-TOPSIS is proposed. Thirdly, a modern intelligent model based on MBA-WLSSVM is designed. In modified bat algorithm (MBA), tent chaotic map is utilized to improve the basic bat algorithm, while weighted least squares support vector machine (WLSSVM) adopts wavelet kernel function to replace the traditional radial basis function to complete the model improvement. Finally, an example is given to verify the scientificity and accuracy of the model, which is helpful for investors to make fast and effective investment risk forecasting of wind energy along the Belt and Road. The example analysis proves that the proposed model can provide reference and basis for investment corpus to formulate the investment strategy in wind energy along the Belt and Road.

A Hyperspectral Image Classification Method Based on Weight Wavelet Kernel Joint Sparse Representation Ensemble and β-Whale Optimization Algorithm

Joint sparse representation (JSR) is a commonly used classifier that recognizes different objects with core features extracted from images. However, the generalization ability is weak for the traditional linear kernel, and the objects with similar feature values associated with different categories are not sufficiently distinguished especially for a hyperspectral image (HSI). In this article, an HSI classification technique based on the weight wavelet kernel JSR ensemble model and the β-whale optimization algorithm is proposed to conduct pixel-level classification, where the wavelet function is acted as the kernel of JSR. Moreover, ensemble learning is used to determine the category label of each sample by comprehensive decision of some subclassifiers, and the β function is utilized to enhance the exploration phase of the whale optimization algorithm and obtain the optimal weight of subclassifiers. Experimental results indicate that the performance of the proposed HSI classification method is better than that of other newly proposed and corresponding approaches, the misclassification and classified noise are eliminated to some extent, and the overall classification accuracy reaches 95% for all HSIs.

Multi-Scale Graph Convolutional Network With Spectral Graph Wavelet Frame

Graph neural networks have achieved impressive progress in solving large and complex graph-structured problems. However, existing methods cannot sufficiently explore the advantages of multi-scale information by enlarging the receptive fields through stacking deep layers or multiplying exponentiated graph Laplacian. They are restricted by the discrete scales rigidly dependent on node-hops or suffer from the over-smoothing issue incurred by long-range low-pass filtering. In this paper, we propose a multi-scale graph convolutional network based on the spectral graph wavelet frame to improve multi-scale representation learning. The proposed network flexibly leverages multi-scale neighboring information with continuous scaling to enhance the discrimination ability of the learned multi-scale representations and offer stable feature extraction ensured by frame bounds in theory. The multi-scale spectral convolutional layer is constructed with a low-pass filter and a sequence of dilated band-pass filters to achieve well-established localization in both vertex and frequency domains. The graph convolutional filters can be realized based on spectral graph wavelet kernel functions analytically formulated using parameterized band-limited filters or adaptively learned from training data. Furthermore, fast approximations of exact spectral filtering are developed for memory efficient operations on sparse tensors with guaranteed prediction accuracy and numerical stability. The proposed network is shown to achieve state-of-the-art performance in extensive experiments on citation networks, bioinformatics graphs, and social networks.

On-chip TIRF nanoscopy by applying Haar wavelet kernel analysis on intensity fluctuations induced by chip illumination.

Photonic-chip based TIRF illumination has been used to demonstrate several on-chip optical nanoscopy methods. The sample is illuminated by the evanescent field generated by the electromagnetic wave modes guided inside the optical waveguide. In addition to the photokinetics of the fluorophores, the waveguide modes can be further exploited for introducing controlled intensity fluctuations for exploitation by techniques such as super-resolution optical fluctuation imaging (SOFI). However, the problem of non-uniform illumination pattern generated by the modes contribute to artifacts in the reconstructed image. To alleviate this problem, we propose to perform Haar wavelet kernel (HAWK) analysis on the original image stack prior to the application of (SOFI). HAWK produces a computational image stack with higher spatio-temporal sparsity than the original stack. In the case of multimoded non-uniform illumination patterns, HAWK processing breaks the mode pattern while introducing spatio-temporal sparsity, thereby differentially affecting the non-uniformity of the illumination. Consequently, this assists nanoscopy methods such as SOFI to better support super-resolution, which is otherwise compromised due to spatial correlation of the mode patterns in the raw image. Furthermore, applying HAWK prior to SOFI alleviates the problem of artifacts due to non-uniform illumination without degrading temporal resolution. Our experimental results demonstrate resolution enhancement as well as reduction in artifacts through the combination of HAWK and SOFI.