Search

Finding patches is a crucial step in a pavement performance survey. The study develops a machine learning and image processing algorithm-established automatic method for identifying asphalt pavement patches. The GrayLevel Co-Occurrence Matrix and image texture-based features derived from color channel statistics are used as input parameters to describe the condition of the pavement. For feature extraction, image processing methods like the projective integral of images, steerable filters, and an improved image thresholding method were used. Support Vector Machine is employed for categorizing the differentiating patched regions from non-patch ones. The suggested combination of image texture analytical methods has been trained using an IA data set created from 400 image samples. The feature set, which combines the characteristics of cracked objects with those generated from the projective integral, can produce the desired result. To make the model's execution simpler, a patch recognition program was created and implemented in MATLAB. As a result, the recently created method has the potential to be an instrument for traffic management organizations through the assessment of pavement performance. The results of the experiments indicate that the newly developed method may achieve excellent accuracy prediction results with a classifier performance rate of about 96%. Such a method could enable transportation authorities to assess the state of asphalt pavement. [ABSTRACT FROM AUTHOR]

Statistical techniques are needed to analyze data structures with complex dependencies such that clinically useful information can be extracted. Individual‐specific networks, which capture dependencies in complex biological systems, are often summarized by graph‐theoretical features. These features, which lend themselves to outcome modeling, can be subject to high variability due to arbitrary decisions in network inference and noise. Correlation‐based adjacency matrices often need to be sparsified before meaningful graph‐theoretical features can be extracted, requiring the data analysts to determine an optimal threshold. To address this issue, we propose to incorporate a flexible weighting function over the full range of possible thresholds to capture the variability of graph‐theoretical features over the threshold domain. The potential of this approach, which extends concepts from functional data analysis to a graph‐theoretical setting, is explored in a plasmode simulation study using real functional magnetic resonance imaging (fMRI) data from the Autism Brain Imaging Data Exchange (ABIDE) Preprocessed initiative. The simulations show that our modeling approach yields accurate estimates of the functional form of the weight function, improves inference efficiency, and achieves a comparable or reduced root mean square prediction error compared to competitor modeling approaches. This assertion holds true in settings where both complex functional forms underlie the outcome‐generating process and a universal threshold value is employed. We demonstrate the practical utility of our approach by using resting‐state fMRI data to predict biological age in children. Our study establishes the flexible modeling approach as a statistically principled, serious competitor to ad‐hoc methods with superior performance. [ABSTRACT FROM AUTHOR]

Performance degradation prediction is considered an effective method to enhance the durability of Proton Exchange Membrane Fuel Cells (PEMFCs). However, accurate prediction under dynamic conditions remains a challenge. The MSD-ICB-LSTM model is proposed to improve the accuracy of the performance degradation prediction, which incorporates the information from both long-term and short-term modes within the PEMFC history operating data. Firstly, redundant variables are removed from the data, and the noise is eliminated using locally weighted scatterplot smoothing (LOWESS) to raise the quality of the input samples. Secondly, the Multi-Scale Decomposition (MSD) method is used to decompose the pre-processed data into long-term and short-term modes. Thirdly, a combined module containing encoding, internal decomposition and a double-layer Isometric Convolution Block (ICB) is used to deeply explore the nonlinear degradation characteristics in the short-term mode, while a Long-Short-Term Memory Network (LSTM) is utilized to estimate the long-term mode. The final result of prediction is obtained by combining the predictions of two modes. The experimental results show that the prediction accuracy of the proposed MSD-ICB-LSTM model is 99.74% in R2，and the Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) are 0.0034 and 0.77%, respectively. Thus, the method can efficiently explore the multi-scale features and enhance the accuracy of PEMFC performance degradation prediction. • An MSD-ICB-LSTM model is proposed to improve the accuracy of PEMFC degradation prediction. • A combined approach of Feature Selection, MI and LOWESS is proposed to eliminate redundant data and noise. • The MSD can extract long-term and short-term modes to distinguish different time scales. • Short-term module with Coding, Internal Decomposition, Two-layer ICB and Merging extract deep nonlinear characteristics. • The long-term module helps gather information over long time scale and balance local overfitting. [ABSTRACT FROM AUTHOR]

Virtual haptic textures can enhance the realism of human–computer interaction, enabling the users in multimedia environments to not only see the virtual objects but also touch the texture details of the objects’ surface. Data-driven approaches have been widely used in haptic texture modeling, but the existing models are either only able to model one texture at a time, or adopt a multimodal approach which increases the modeling complexity. This article presents a method for implementing a cross-texture haptic model capable of generating tactile vibration according to the given texture type and user’s motion. The modeling method is based on tactile feature fusion, primarily divided into two steps. First, hand-crafted texture features, motion features and vibration features are extracted from the pre-collected raw tactile data. These features are interpretable and have their own physical meanings. Then, a neural network with fully connected layers is devised for its non-linear advantage in simulating vibration response and trained to fuse texture feature and motion feature before mapping them to vibration feature. The proposed model is validated through objective and subjective experiments using two datasets. The experimental results demonstrate its accuracy in generating numerical signals and its capability to run in real-time. [ABSTRACT FROM AUTHOR]

Internet of Things (IoTs) has been playing an important role in many sectors, such as smart cities, smart agriculture, smart healthcare, and smart manufacturing. However, IoT devices are highly vulnerable to cyber-attacks, which may result in security breaches and data leakages. To effectively prevent these attacks, a variety of machine learning-based network intrusion detection methods for IoT networks have been developed, which often rely on either feature extraction or feature selection techniques for reducing the dimension of input data before being fed into machine learning models. This aims to make the detection complexity low enough for real-time operations, which is particularly vital in any intrusion detection systems. This paper provides a comprehensive comparison between these two feature reduction methods of intrusion detection in terms of various performance metrics, namely, precision rate, recall rate, detection accuracy, as well as runtime complexity, in the presence of the modern UNSW-NB15 dataset as well as both binary and multiclass classification. For example, in general, the feature selection method not only provides better detection performance but also lower training and inference time compared to its feature extraction counterpart, especially when the number of reduced features K increases. However, the feature extraction method is much more reliable than its selection counterpart, particularly when K is very small, such as K = 4 . Additionally, feature extraction is less sensitive to changing the number of reduced features K than feature selection, and this holds true for both binary and multiclass classifications. Based on this comparison, we provide a useful guideline for selecting a suitable intrusion detection type for each specific scenario, as detailed in Table 14 at the end of Sect. 4. Note that such the comparison between feature selection and feature extraction over UNSW-NB15 as well as theoretical guideline have been overlooked in the literature. [ABSTRACT FROM AUTHOR]

This research paper aims to ensure the reliable and efficient operation of grid-connected solid-state transformers (SSTs) by detecting and evaluating various undesirable operating conditions. The study considers different types of faults, including internal faults like open switches and open capacitors, external faults such as symmetrical and asymmetrical faults occurring at various locations of the SST, and abnormalities on the grid side known as sympathetic inrush conditions. To analyze these operating conditions, the secondary current of the high-frequency transformer is normalized and decomposed using the discrete wavelet transform (DWT) and wavelet packet transform (WPT). From the DWT and WPT decomposition at multiple levels, several statistical parameters are calculated. These statistical parameters are carefully selected from different decomposition levels to enhance the effectiveness of the detection algorithm utilizing DWT and WPT. In order to quickly identify and classify all operating conditions that impact the performance of the grid-connected SST, a three-layer feedforward artificial neural network (ANN) is employed, using the selected statistical features. The accuracy and efficiency of the ANN-based classification approach are evaluated by assessing the effectiveness of the statistical features obtained from DWT and WPT. Simulation results have been altered by introducing various noise levels to systematically assess the performance of the proposed algorithms. The average accuracy of the DWT-ANN algorithm stands at 97.89%, while the WPT-ANN algorithm achieves a slightly elevated accuracy level of 98.01%. This notable similarity in accuracy curtails from the careful selection of the wavelet function, decomposition level, and feature sets. [ABSTRACT FROM AUTHOR]

Smart contracts function like specialized computer programs on the blockchain. Many of these contracts are on Ethereum, but sometimes these contracts have problems with security. These problems caused big money losses and made the blockchain less stable. Smart contracts are self-executing with predefined rules and are at the core of many blockchain applications. However, they are susceptible to various vulnerabilities and security risks. Automated vulnerability detection helps identify and mitigate these issues efficiently. Smart Contracts (SC) have become really popular lately. People think they are the future for making deals on blockchains. Smart contracts are like automatic agreements. They work by themselves using special computer programs. They follow the rules of the deal and keep track of everything. The main idea with smart contracts is to get rid of the need for traditional trusted middlemen-like authorities or organizations. Instead, we use code that runs on a secure and unchangeable system. In this manuscript, pioneering automated vulnerability detection for smart contracts in blockchain using KEVM: Guardian ADRGAN (ADRGAN-SCB-KEVM) is proposed. Here, K framework's Ethereum virtual machine (KEVM) is a computation engine used in this research. From this KEVM, smart contracts data are provided to feature extraction phase. Feature extraction is done using Nested patch-based feature extraction. Then the extracted features are fed to attentive dual residual generative adversarial network (ADRGAN), identifying KEVM smart contracts vulnerabilities. Finally, by using ADRGAN, it is classified as Vulnerable and Non-Vulnerable in smart contracts. The proposed ADRGAN-SCB-KEVM method employed on Python and efficiency of proposed method evaluated with different metrics like Accuracy, Computation Time, Precision, Recall, F1 Score, Specificity, RoC are evaluated. The simulation outcomes prove that the proposed ADRGAN-SCB-KEVM technique attains 41.34%, 31.28%, and 36.38% higher Accuracy for Vulnerable; 32.44%, 38.45%, and 29.47% higher Accuracy for Non-Vulnerable while compared with the existing methods such as Utilizing fault injection to evaluate blockchain systems in the presence of faulty smart contracts (UFI-BS-ESVD), State-of-the-Art Blockchain-Enabled Smart Contract Applications in the University (SA-BESC-ESVD), enhancing Ethereum smart contracts static analysis by computing precise Control-Flow Graph of Ethereum bytecode (EESC-CPC-EBD), respectively. [ABSTRACT FROM AUTHOR]

The rapid growth of the Internet of things (IoT) platform has implications on security vulnerabilities that need to be resolved. This requires an intrusion detection system (IDS) to secure attacks on the platforms. In line with this, numerous machine and deep learning algorithms have been adopted to detect cyber-attacks. Real-time IoT devices transmit massive amounts of heterogeneous data, which affects the network. Traffic networks generate redundant and large amounts of data that must be reduced before processing. This study proposed a hybrid deep learning model for an IDS on the IoT platform. We used unsupervised approaches to extract data dimensions and features, then a neural network for classification. Several approaches were used to determine the effectiveness of the deep learning-based IoT IDS with two scenarios of feature extraction. The first case used autoencoder variants such as deep autoencoder (DAE), deep LSTM autoencoder (LSTM-DAE), and deep convolutional autoencoder. The second case used stacked models for feature extraction, including stacked autoencoder and deep belief network. The feature extraction output from the five models was fine-tuned to the fully connected layer using the BoT-IoT dataset. The results showed a good detection performance of almost 100% and a false positive rate (FPR) of nearly 0%. On the CSE-CIC-IDS2018 dataset, the proposed deep learning model was evaluated using a transfer learning approach with the highest detection rate of 99.17% and the lowest FPR of 0.18%. The model developed from the feature extraction process recognized attacks significantly better than the previous approach. [ABSTRACT FROM AUTHOR]

Aiming to solve the problems of multiple internal power components, high fault probability, high similarity of the fault features of different power components, difficulty of traditional fault diagnosis feature extraction and low accuracy of fault identification in high-voltage multilevel cascaded H-bridge inverters, this paper presents a fault diagnosis method based on an attention-parallel multisource data fusion residual network. First, a parallel residual neural network model is established, and the extracted multilevel three-phase voltage before filtering and the three-phase current waveform after filtering are converted into two-dimensional image data using a wavelet transform. Subsequently, a feature fusion module is integrated into the network structure to adaptively extract features at different network levels. This module locates key features using the attention mechanism. Then, it fuses useful fault information into feature images using the feature fusion mechanism, enhancing the feature representation capability of the network. Finally, the fault features extracted by the feature fusion module undergo the complete convolution operation. The final enhanced features are used as classification features and classified using a softmax layer. Experimental results demonstrate that the proposed method exhibits high fault diagnosis accuracy and adaptability. [ABSTRACT FROM AUTHOR]

The development of geospatial artificial intelligence (GeoAI) systems depends on the ability to learn effective representations of places. To learn accurate place representations from spatial interactions, it is important to extract features that capture both the spatial and non-spatial driving factors. However, existing methods lack a robust interpretation and the explanatory power of the learned representations on spatial factors remains unexplored. Here, we propose an approach to learning place representations from spatial interactions. Our method is inspired by flow allocation, which is the main focus of single-constrained gravity models. We first validate the method on synthetic flows with known driving factors and then apply it to multi-scale real-world flows. Results show that the learned representations can effectively capture features that explain place characteristics, along with the impact of spatial impedance. Our study not only contributes an efficient method to learn place representations from spatial interactions but also offers insights into pre-training procedures in GeoAI. [ABSTRACT FROM AUTHOR]

The presence of road cracks is an important indicator of damage. Deep learning is a prevailing method for detecting cracks in road surface images because of its detection ability. Previous research works focused on supervised convolutional neural networks (CNNs) without non‐crack features or unsupervised crack analysis with limited accuracies. The novelty of this study is the addition of background classification. By increasing the number of non‐crack categories, CNNs are driven to learn non‐crack features and improve crack detection performances. Non‐crack images are preprocessed, and their features are extracted in an unsupervised way by a deep convolutional autoencoder. A self‐organizing map clusters features to obtain non‐crack categories. This study focusses on classification though the method can be adopted in parallel with the latest segmentation algorithms. Using common road crack datasets, modified deep CNN models significantly improved accuracy by 1%–4% and f‐measure by 3%–8%, compared to previous models. The modified visual geometry group (VGG) 16 showed the top‐level performance, 96% accuracy and 84%–85% f‐measure. The models drastically reduced false detection cases while maintaining their crack detection abilities. [ABSTRACT FROM AUTHOR]

Motor imagery electroencephalogram (MI-EEG) is becoming increasingly important. This paper solves the problem of online signal recognition for motor imagery across subjects by finding common features across multiple subjects to improve the generality of the classification model. We analysed the EEG data from left/right-hand motor imagery of eight subjects and proposed a weighted time-domain (WTD) feature extraction method based on a weighted channel screening method. The classification model constructed by combining this feature extraction method with the support vector machine (SVM) classification method was faster in classification and achieved good cross-subject classification accuracy (The average offline classification accuracy was 91.39%). In this paper, an online control system for asynchronous brain-controlled wheelchairs was built with good performance. The online average motor imagery classification accuracy was 81.67%, and the average response time was 1.36s. This method contributes to bringing the online Brain-computer interface (BCI) system out of the laboratory and into wider application. [ABSTRACT FROM AUTHOR]

At present, the majority of techniques for assessing image quality are limited to extracting features from an image in a single space. This paper proposes a new dual-space multi-feature fusion based method for full-reference image quality assessment. This method involves simultaneously extracting features from both the YIQ and L*a*b* color spaces. First, we extract the luminance, slope, chroma, and gradient features in the spatial domain of the image to describe the salient differences in the image. Second, based on contrast sensitivity characteristics, we extract spatial frequency features in the spatial domain of the image to represent frequency differences in the image. Next, merge the features extracted in the dual space to construct a quality perception feature vector. Finally, the feature vector is input into the Random Forest model for regression prediction to obtain the predicted score of the image. Many experiments have been carried out on the four public datasets, and contrasted with other methods. The experimental confirm that the proposed method predicts image quality more accurately. The MATLAB source code and dataset of this paper will be published on GitHub, and the corresponding author can be contacted if necessary. [ABSTRACT FROM AUTHOR]

• A random forest model is proposed for cell-type-specific cohesin-mediated chromatin loops prediction. • Chromatin state is responsible for cell-type-specificity of chromatin loops. • Chromatin state combines with the binding of key factors can predict cell-type-specific loops accurately. Chromatin loop is of crucial importance for the regulation of gene transcription. Cohesin is a type of chromatin-associated protein that mediates the interaction of chromatin through the loop extrusion. Cohesin-mediated chromatin interactions have strong cell-type specificity, posing a challenge for predicting chromatin loops. Existing computational methods perform poorly in predicting cell-type-specific chromatin loops. To address this issue, we propose a random forest model to predict cell-type-specific cohesin-mediated chromatin loops based on chromatin states identified by ChromHMM and the occupancy of related factors. Our results show that chromatin state is responsible for cell-type-specificity of loops. Using only chromatin states as features, the model achieved high accuracy in predicting cell-type-specific loops between two cell types and can be applied to different cell types. Furthermore, when chromatin states are combined with the occurrence frequency of CTCF, RAD21, YY1, and H3K27ac ChIP-seq peaks, more accurate prediction can be achieved. Our feature extraction method provides novel insights into predicting cell-type-specific chromatin loops and reveals the relationship between chromatin state and chromatin loop formation. [ABSTRACT FROM AUTHOR]

• Representing features at multiple scales is of great importance for numerous vision tasks. • Recent advances in backbone convolutional neural networks (CNNs) continually demonstrate stronger multi-scale representation ability, leading to consistent performance gains on a wide range of applications. • Most existing methods represent the multi-scale features in a layer-wise manner. • Res2Net has hierarchical residual-like connections within one single residual block. • Res2Net represents multi-scale features at a granular level and increases the range of receptive fields for each network layer. Obtaining an accurate segmentation of the pulmonary nodules in computed tomography (CT) images is challenging. This is due to: (1) the heterogeneous nature of the lung nodules; (2) comparable visual characteristics between the nodules and their surroundings. A robust multi-scale feature extraction mechanism that can effectively obtain multi-scale representations at a granular level can improve segmentation accuracy. As the most commonly used network in lung nodule segmentation, UNet, its variants, and other image segmentation methods lack this robust feature extraction mechanism. In this study, we propose a multi-stride residual 3D UNet (MRUNet-3D) to improve the segmentation accuracy of lung nodules in CT images. It incorporates a multi-slide Res2Net block (MSR), which replaces the simple sequence of convolution layers in each encoder stage to effectively extract multi-scale features at a granular level from different receptive fields and resolutions while conserving the strengths of 3D UNet. The proposed method has been extensively evaluated on the publicly available LUNA16 dataset. Experimental results show that it achieves competitive segmentation performance with an average dice similarity coefficient of 83.47 % and an average surface distance of 0.35 mm on the dataset. More notably, our method has proven to be robust to the heterogeneity of lung nodules. It has also proven to perform better at segmenting small lung nodules. Ablation studies have shown that the proposed MSR and RFIA modules are fundamental to improving the performance of the proposed model. [ABSTRACT FROM AUTHOR]

• Protein lysine methylation plays an important role in proteins' function regulation. • Here we propose CNN-Meth to accurately predict lysine methylation sites. • To build CNN-Meth, we use convolutional neural network (CNN). • CNN-Meth is trained using evolutionary, structural, and physicochemical features. • CNN-Meth and its source code are publicly available at https://github.com/MLBC-lab/CNN-Meth. Protein lysine methylation is a particular type of post translational modification that plays an important role in both histone and non-histone function regulation in proteins. Deregulation caused by lysine methyltransferases has been identified as the cause of several diseases including cancer as well as both mental and developmental disorders. Identifying lysine methylation sites is a critical step in both early diagnosis and drug design. This study proposes a new Machine Learning method called CNN-Meth for predicting lysine methylation sites using a convolutional neural network (CNN). Our model is trained using evolutionary, structural, and physicochemical-based presentation along with binary encoding. Unlike previous studies, instead of extracting handcrafted features, we use CNN to automatically extract features from different presentations of amino acids to avoid information loss. Automated feature extraction from these representations of amino acids as well as CNN as a classifier have never been used for this problem. Our results demonstrate that CNN-Meth can significantly outperform previous methods for predicting methylation sites. It achieves 96.0%, 85.1%, 96.4%, and 0.65 in terms of Accuracy, Sensitivity, Specificity, and Matthew's Correlation Coefficient (MCC), respectively. CNN-Meth and its source code are publicly available at https://github.com/MLBC-lab/CNN-Meth. [ABSTRACT FROM AUTHOR]

As one of the key technologies in battery management system, accurate remaining useful life (RUL) prediction is critical to guarantee the reliability and safety for electrical equipment. However, the generalization and robustness of a single method are limited. A novel fusion data‐driven RUL prediction method CSSA‐ELM‐LSSVR based on charging‐discharging health features extraction is proposed in this paper, which fusions chaotic sparrow search algorithm (CSSA), extreme learning machine (ELM), and least squares support vector regression (LSSVR). First, four health indicators (HIs) are extracted from the charging‐discharging process, which can reflect the battery degradation phenomenon from multiple perspectives. Then, pearson correlation coefficient is used to numerically analyze the correlation between HIs and battery aging capacities. Second, the extracted HIs are used as the inputs for ELM and LSSVR to predict the degradation trend of battery, where CSSA is used for hyperparameters optimization in ELM. Finally, considering that CSSA‐ELM can capture the general trend of degradation curves, while LSSVR can trace the detail changes, a fusion framework based on CSSA‐ELM and LSSVR is proposed for RUL prediction. Two weighting schemes, namely precision‐based weighting (PW) and random forest regressor‐based weighting (RFRW) are put forward to fix the weights of CSSA‐ELM and LSSVR algorithms. Two publicly available datasets from National Aeronautics and Space Administration (NASA) and MIT are adopted to verify the feasibility and effectiveness of the proposed method. The results indicate that the proposed method with any weighting scheme has an overall superior prediction performance for different kinds of batteries compared with CSSA‐ELM, LSSVR, convolution neural network and long short term memory. Moreover, the RFRW scheme has better overall performance. Specifically, the maximum root mean square error of the predicted method is 2.5126%, the mean absolute percentage error is 12.9167%, the mean absolute error is 1.8376%, the predicted RUL errors are within one cycle, and the determination coefficient R2$R^2$ is above 0.97. [ABSTRACT FROM AUTHOR]

Predicting the remaining useful life (RUL) of an engine is one of the key tasks of Prognostics and health management (PHM). Modern mechanical equipment typically operates in complex operating conditions and fault modes, leading to dispersed distribution of sensor data and challenges for feature extraction. To improve the accuracy of predicting the RUL under the complex scenarios, this paper proposes a multi‐scale convolutional network (CNN) and bidirectional gated recurrent unit (MSC‐BiGRU) mode under a dual path framework with temporal attention. Specifically, the multi‐scale CNN in the first path is to learn complex features, and Swish Activation function is used to improve the prediction ability of the network; the bidirectional gated recurrent unit (BiGRU) in the second path can handle both forward and backward time series, and adaptively capture the importance of outputs at different times using temporal attention, enhancing the model's feature extraction ability in the temporal dimension. A feature fusion mechanism is developed to connect two paths in parallel, overcoming the overfitting and high computational complexity in deep complex models. We verify the effectiveness of the proposed method using a simulated turbofan engine dataset, especially on datasets FD002 and FD004 under complex operating conditions and fault modes, the RMSE values were reduced by 17.37% and 9.97%, respectively, compared to the BiGRU‐TSAM. [ABSTRACT FROM AUTHOR]

Nowadays, computational intelligence-assisted autonomous assessment of education quality has become a more and more general concern in the area of smart education management. As education quality assessment is a complicated process with multiple heterogeneous factors, it remains challenging to make effective assessment using simple information modality and criteria. To deal with this issue, this paper introduces multiscale deep learning, and proposes a novel data-driven autonomous assessment framework for education quality. In particular, four aspects of heterogeneous indicators are selected as the basic indexes and a dilated convolutional neural network structure is formulated to perform multiscale feature extraction. Then, the structural equation is adopted to make multivariate characterization and output the final assessment results. At last, some simulations are carried out on realistic education operation data to evaluate the efficiency of the proposed autonomous assessment framework. Two aspects of findings can be deduced from the results. For one thing, multisource fusion of multiple indicators well makes sense to autonomous education assessment. For another, multiscale deep learning can provide some beneficial promotion to assessment efficiency. [ABSTRACT FROM AUTHOR]

This study proposes a fatigue life prediction method for composite bolted joints, which combines algorithm optimization‐based hybrid neural networks with finite element modeling. First, based on the Hashin failure criterion of physical mechanism, a finite element model for fatigue life prediction of composite bolted joints is established, and the simulation calculations have been conducted using various initial conditions. Then, by integrating the simulation and experiment data, we have established a fatigue life database that serves machine learning training and prediction. Finally, the data undergo a comprehensive process of deep feature extraction through the utilization of a convolutional neural network (CNN). The resulting deep features are utilized as inputs for training the backpropagation neural network (BPNN) to predict fatigue life. The results indicate this synergistic combination of CNN and BPNN results in a substantial improvement in prediction accuracy and has remarkable superiority in predicting the fatigue life of composite bolted joints. Highlights: An FE model is established to predict the fatigue life of composite bolt joints.A fatigue life prediction model is proposed based on the hybrid neural network.The advantages of convolution neural network and BP neural network are combined.The superiority of the CNN–BP hybrid neural network has been verified. [ABSTRACT FROM AUTHOR]

There exist numerous multi‐path signals with impulsive noise (IN) in a multi‐path propagation environment. The direction‐of‐arrival (DOA) is highly challenging to be assessed because of the strong IN. For various military along with civilian applications, DOA estimation has turned into a hopeful technology. But, estimating DOA for multipath signals with IN environments is extremely complicated. A desirable output is not attained by several existent techniques, namely subspace‐centric approaches, maximum likelihood techniques, along with sparse representation‐centric methods, for DOA assessment. Therefore, an effective satin bowerbird optimization‐based deep learning neural network centered DOA estimation upon single snapshots is proposed in this paper for attaining robust and precise DOA estimation in multipath and IN environments. An arbitrary array configuration is employed by utilizing the proposed work as an input signal for establishing the received signal's model. Pre‐processing, time‐frequency domain conversion, feature extraction, feature reduction and DOA estimation are the proposed method's 5 disparate stages. To evince the proposed model's efficacy, its outcomes are analogized with other prevailing techniques regarding some performance metrics. The proposed model attains the root mean square error of 0.6484366‐AZ and 0.6484366‐EA, root mean squared percentage error (RMPSE) of 4.636107‐AZ and 5.628599‐EA and probability of resolution of 0.967999‐AZ and 0.962841‐EA, which are extremely less when compared to the prevailing methods; thus, depicting the superiority of the proposed model for an effective DOA estimation. [ABSTRACT FROM AUTHOR]

Analysis of video sequences of public places is an important topic in video surveillance systems. Due to the high probability of occurring abnormal behavior in crowded scene, the main purpose of many surveillance systems is to monitor the crowd movement, and detection of abnormalities. To speed up this process and also for error reduction, it is highly important to use automated and intelligent tools in surveillance systems, as an alternative to the human operator. This study presents an unsupervised and online algorithm for analysis of dynamic crowd behavior, which uses the proposed features, with the capability to analyze crowds over time and reveal different behaviors of the crowd groups. In the proposed algorithm, prominent points are initially tracked. These key points are processed by the proposed system that includes removing the fixed points, employing proposed features of the moving points, automated determination of neighborhood, the similarity of the invariant neighbors. Group clustering is done automatically and the classification stage is conducted without the training phase. The dynamic behavior of the crowd is examined using the features and the extracted group properties and different states in the scene are diagnosed by dynamic thresholding. Experimental evaluation of the proposed method on several databases shows that it is performed properly in video sequences and it is able to detect various abnormal behaviors in the crowd scenes. [ABSTRACT FROM AUTHOR]

Over the past few decades, irrigation using groundwater has increased significantly. It has significant effects on local to regional climates as well as terrestrial energy fluxes, food production, and water availability. High cost of metering equipment installation as well as maintenance, privacy concerns, and existence of unregistered or illegal wells make it difficult to monitor irrigation water use on a large scale. This study suggests a unique approach to DL-based feature extraction and categorization for ecosystem-based water management in agricultural fields. Agriculture field water analysis data were used as the input in this instance, which was subsequently processed for noise removal, smoothing, and normalisation. Particle swarm-based convolutional architecture has been used to extract the processed data feature. Back regressive propagation based on incentive Q-learning is used to classify the extracted features. Experimental analysis has been carried out in terms of accuracy, precision, recall, F-1 score, RMSE and mAPE. Proposed technique obtained accuracy of 92%, precision of 78%, recall of 83%, F_1 score of 76%, RMSE of 55% and MAPE of 57%. [ABSTRACT FROM AUTHOR]

For irrigation in agriculture, water is a natural resource. Recycling water use is vital for the sustainable development of ecological environment and for resource conservation. Different substances that are thought to be pollutants and contribute to the deterioration of water quality are present in the wastewater from daily life and industrial activity. This research propose novel method in agricultural water management using feature extraction as well as classification based on DL methods. Inputs are collected as agriculture field water management as well as processed for noise removal, normalization and smoothening. Processed input data features are extracted utilizing kernel convolutional component analysis network. The extracted features has been classified using Quadratic reinforcement NN. Experimental analysis are carried out in terms of accuracy, precision, recall, positive predictive value, RMSE and mAP. Proposed technique attained accuracy of 92%, precision of 86%, recall of 65%, positive predictive value of 71%, RMSE of 55%, MAP of 51%. [ABSTRACT FROM AUTHOR]

To explore the potential of utilising radiomics analysis and machine-learning models that incorporate intratumoural and peritumoural regions of interest (ROIs) for predicting brain metastasis (BM) in newly diagnosed lung cancer patients. The study comprised 183 lung cancer patients (training cohort: n= 146; validation cohort: n= 37) whose radiomics features were extracted from plain computed tomography (CT) images of the primary lesion. Four machine-learning algorithms (logistic regression [LR], support vector machine [SVM], k-nearest neighbour algorithm [KNN], and random forest [RF]) were employed to develop predictive models. Model diagnostic performance was assessed through receiver operating characteristic (ROC) analysis, and clinical utility was evaluated using decision curve analysis (DCA). Finally, the radiomics model's generalisation ability was further validated in the prediction of metachronous brain metastasis (MBM). After feature screening, 22 radiomics features were identified as highly predictive, of which nine were derived from the peritumour region. All four machine-learning models demonstrated predictive capability, with SVM showing superior efficiency and robustness. The area under the ROC curve (AUC) of SVM was 0.918 in the training cohort and 0.901 in the validation cohort. DCA indicated the highest net benefit. Furthermore, the time-dependent ROC curve exhibited predictive efficacy for MBM occurrence across 1-, 2-, and 3-year follow-up periods, with all AUC values exceeding 0.7. The optimal SVM model integrating intratumoural and peritumoural radiomics features was confirmed and defined as an imaging biomarker for predicting BM in newly diagnosed lung cancer patients, underscoring its potential to significantly impact clinical diagnosis and treatment. • 22 radiomic features identified as predictors. • Developed four effective machine learning models. • SVM optimally predicts initial brain metastasis. • SVM also forecasts subsequent brain metastasis. [ABSTRACT FROM AUTHOR]

The area of the brain affected by a brain tumour can be identified using the tumour's shape, size, location, and border. This study seeks to develop a novel system of classification for brain tumours through pre-processing, segmentation, feature extraction, and tumour classification. An improved median filter will be applied to the input image in this initial phase to improve it. In this step, the image is segmented using a U-net model. Then, characteristics based on the Median Binary Pattern (MBP), the loop, the modified Local Gabor Directional Pattern (LGDiP), and the tumour size are retrieved. A hybrid model that fuses DBN and Bi-LSTM is presented to classify cancers. The optimal weights for both classifiers will be tuned during training to improve the classification performance. For this, BMEBEO (Blue Monkey Extended Bald Eagle Optimization) is proposed, which is a hybrid optimization technique. The suggested model obtains the maximum F-measure of 96.16%%. [ABSTRACT FROM AUTHOR]

The magnetic flux leakage (MFL) technique is widely employed for nondestructive testing of ferromagnetic specimens and materials, including wire ropes, bridge cables, and pipelines. As regards the MFL testing, extracting features from MFL signals is crucial for defect recognition and estimation of corresponding widths. Deep learning has been extensively used for feature extraction, but it often performs inadequately on a small sample dataset. To address this limitation, this paper develops a network framework that combines the Wavelet Scattering Transform (WST) and Neural Networks (NN) for defect width estimation. The WST is a knowledge-based feature extraction technique with a structure similar to convolutional neural networks. It offers a translation-invariant representation of signal features using a redundant dictionary of wavelets. The NN then maps the WST feature representation to the defect width information. Experiments on real steel plates with defects are carried out to validate the effectiveness of the proposed framework. Quantitative comparisons of the experimental results demonstrate that the proposed framework achieves better estimation performance in handling MFL signals and has superiority in scenarios with limited training samples. [ABSTRACT FROM AUTHOR]

This study assessed the viability of using hyperspectral imaging (HSI) technology for nondestructive detection of moisture content in oilseed rape leaves. Besides, a method (IVISSA-iPLS) coupling interval variable iterative space shrinkage approach (IVISSA) with interval partial least square (iPLS) was introduced to identify characteristic wavelengths. The IVISSA-iPLS algorithm changed the selection target from wavelength points to spectral intervals, reducing the computational burden while increasing the continuity between the selected wavelengths. Subsequently, the characteristic wavelengths selected by the IVISSA-iPLS were used as the input of the least square support vector regression (LSSVR) model to predict the moisture content of oilseed rape leaves. Additionally, the competitive adaptive reweighted sampling (CARS), the successive projections algorithm (SPA), the IVISSA, and the iPLS were investigated as wavelength selection algorithms for comparison. The results indicated that the LSSVR models based on the characteristic wavelengths acquired from the IVISSA-iPLS using divided wavelength intervals of 30, demonstrated the highest performance, with R p 2 of 0.9555, RMSEP of 0.0065, and RPD of 4.715. Finally, the optimal prediction model was used to visualize the moisture content of oilseed rape leaves, which offered a more intuitive and effective method for the evaluation of moisture content. The results ascertained the significant possibility of combining HSI with combinatorial algorithms in detecting, quantifying, and visualizing the moisture content of oilseed rape leaves. [ABSTRACT FROM AUTHOR]

The response of a nonlinear system excited by multi-frequency signals and Poisson white noise contains abundant dynamic behaviours. The study on system response has certain practical values. Utilising optimal dynamics response has unique advantages in weak feature identification under strong background noise. In this work, we study the optimal dynamics response in the bistable system excited by multi-frequency signals and Poisson white noise. The effects of noise parameters on system response and feature identification are discussed and analysed. Also, a signal identification method for unknown multi-frequency signals based on the optimal resonance response of the nonlinear system is proposed and applied to the compound fault diagnosis of rolling bearings, which utilises the difference between stochastic resonance and coherence resonance to determine characteristic components of the signals. The proposed method effectively identifies the characteristic components of the original excitation and avoids misjudgment caused by strong noise, which is effectively verified via simulation analysis and experiments. [ABSTRACT FROM AUTHOR]

Environmental applications play a significant role in the ongoing research area of Polarimetric Synthetic Aperture Radar (PolSAR) image classification. In this paper, a new model is proposed for classifying PolSAR images and applied to a part of the Greater Cairo area in the Nile basin, South of Delta, Egypt. First, the proposed model performs data pre-processing by extracting the coherency and covariance elements noted as [T] and [C] matrices, respectively. Second, temporal convolutional networks (TCN) deep learning is used to extract the features from coherency and covariance elements and then train the model. Third, the SoftMax classifier is used to classify the PolSAR image. Finally, the proposed model is tested with evaluation metrics. The obtained results show that the proposed model can achieve high classification performances. [ABSTRACT FROM AUTHOR]

Complex risk factors make metro construction safety accidents prone to occur, and there are various types of accidents. Accident reports record detailed information about different types of accidents in text form. However, effectively utilizing such unstructured data presents a significant challenge. Text mining (TM) provides a viable foundation for addressing this challenge, but related studies have limitations in risk feature extraction and lack of in-depth analysis capability. To address the deficiencies of existing studies and provide a feasible strategy for identifying key risk factors in the metro construction domain, this paper proposes an integrated model combining TM and machine learning–based Bayesian networks. Firstly, the term frequency-inverse document frequency (TF-IDF) algorithm in TM was used to separately extract the direct and indirect cause factors from the accident reports, with the missing factors supplemented using the TextRank algorithm. Then, depending on the assumption of whether to consider the conditional independence between factors, an improved naive Bayesian network (NBN) and a tree-augmented naive Bayesian network (TAN) were built based on the extracted factors and the corresponding accident types, respectively, for further in-depth analysis. Finally, the training set was divided to train the two network models, and sensitivity analysis was used to identify the key risk factors. Using 162 accident reports from China as an application example, the results showed that TAN exhibited a higher average accuracy (79.62%) in the test set compared with the improved NBN (71.75%), and the importance of risk factors for different accident types was successfully ranked from multiple perspectives using TAN. Meanwhile, some new insights into metro accidents in China were obtained, which can support decision-making for accident prevention and control. In conclusion, this paper effectively addresses the relevant limitations of accident text utilization and presents a novel approach for metro construction safety management. Analyzing accident texts can help gain insights from objective historical data to support safety management efforts. However, accident texts are often unstructured and contain a lot of irrelevant content. How to quickly extract valid information from accident text and use it to analyze accidents in depth is of continuous interest to safety managers. In particular, those models that have real-time decision support capabilities in addition to theoretical insights. This paper proposes an integrated model that combines text mining and machine-learning Bayesian networks. This model achieves comprehensive textual feature extraction, multifaceted accident causation analysis, and allows safety managers to input current accident information into the model to obtain real-time decision support for accident prevention and control. Although the proposed model is developed for metro construction, it can be slightly adapted by incorporating the characteristics of accident texts from similar domains to obtain an integrated model suitable for these domains, so as to effectively control the occurrence of safety accidents. [ABSTRACT FROM AUTHOR]