Search

• Qualitative and quantitative detection of amine gases could be achieved by CSA. • A visible detection of beef freshness using the amine-responsive CSA was proposed. • ResNet34 had the best performance for beef freshness detection based on CSA. • T-SNE could further visualize and understand the classification process of DL. This study developed an innovative approach that combines a colourimetric sensor array (CSA) composed of twelve pH-response dyes with advanced algorithms, aiming to detect amine gases and assess the freshness of chilled beef. With the assistance of multivariate statistical analysis, the sensor array can effectively distinguish five amine gases and enable rapid quantification of trimethylamine vapour with a limit of detection (LOD) of 8.02 ppb and visually monitor the fresh levels of chilled beef. Moreover, the utilization of deep learning models (ResNet34, VGG16, and GoogleNet) for chilled beef freshness evaluation achieved an overall accuracy of 98.0 %. Furthermore, t -distributed stochastic neighbour embedding (t -SNE) visualized the feature extraction process and provided explanations to understand the classification process of deep learning. The results demonstrated that applying deep learning techniques in the process of pattern recognition of CSA can help in realizing the rapid, robust, and accurate assessment of chilled beef freshness. [ABSTRACT FROM AUTHOR]

Offshore wind-hydrogen systems operate in harsh marine environments for extended periods, posing risks of low accessibility and high failure rates. This paper proposes a data-driven fault detection framework for offshore wind-hydrogen systems, aiming to promptly detect potential faults based on the system's physical parameters. The research establishes a comprehensive model based on the operating principles of offshore wind-hydrogen systems. The study uses fault injection techniques to simulate common partial faults in offshore wind-hydrogen systems and collects relevant physical parameters (such as wind turbine power generation, energy storage battery charging/discharging current, and electrolyzer hydrogen production rate). A fault detection model was constructed using a combination of Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) to enable precise detection of potential faults in the system. The established model provides theoretical support for designing, optimizing, and operating offshore wind-hydrogen systems, reducing reliance on actual experiments and lowering project risks and costs. Introducing artificial faults can evaluate the system's performance in different fault scenarios. This paper proposes a CNN-BiLSTM fault detection method that achieves automatic feature extraction and high-precision classification of time-series fault data, confirming the practicality and feasibility of offshore wind-hydrogen systems. This investigation realizes the integration of the offshore wind-hydrogen system model and deep learning, as well as a fully automated learning process from data to fault detection. By calculating and comparing results, it is verified that the framework has a high accuracy of 96.9%. This innovative research offers new methods and perspectives for enhancing the reliability and efficiency of fault detection in offshore wind-hydrogen systems. • A comprehensive multi-system simulation model for offshore wind-hydrogen systems is established. • Fault injection is employed to acquire data samples for offshore wind-hydrogen systems. • Utilizing the CNN-BiLSTM method, a high-accuracy deep learning fault detection model is crafted. • A simulation model-based data-driven deep learning framework for fault detection is devised in this study. [ABSTRACT FROM AUTHOR]

Breast cancer (BC) continues to be a prominent issue in global public health, emphasizing the need for precise and timely detection. This paper employs a deep learning (DL) approach to introduce an extensive methodology for categorizing histopathology images associated with breast cancer into automated binary classifications. The proposed framework architecture is validated on the standard database which is accessible to the public, called Breast Cancer Histopathological Database (BreakHis). Data augmentation techniques are employed for the pre-processing stage. This paper uses the DenseNet 121 pre-trained model for feature extraction and fully connected layers (FCL) to fine-tune the model further. In this experiment, the highest accuracy of 96.09% is observed with the 100X. The experimental results showed an improvement in accuracy for all the magnification factors compared to the existing works. [ABSTRACT FROM AUTHOR]

Human activity recognition(HAR) is an important research focus in ubiquitous computing. It has been widely applied in various domains, such as smart homes, healthcare assistance, and sports training. Accurate human activity recognition directly impacts the performance of downstream tasks. Existing methods for human activity recognition primarily rely on extracting temporal or spatial features from the data. These features are used for the task of human activity recognition. The existing methods mainly use CNN or RNN models to model the Euclidean correlations among spatially adjacent sensors or channels. However, non-Euclidean pairwise correlations among all sensors or channels are even critical for accurate classification, which has been ignored by the existing methods. In this paper, we incorporate fine-grained spatial structural data into the model to overcome these limitations. A novel deep learning model for human activity recognition is proposed, which is called the fine-grained data-oriented Spatial-Temporal Graph Transformer network (STGT). The introduction of the STGT model can eliminate the limitations of existing spatial feature extraction methods by leveraging a novel data organization approach proposed in this study. This model enhances the spatial features within the time feature extraction module for effective spatio-temporal feature extraction. We conducted experiments on four large-scale real-world HAR datasets to evaluate its performance. The experimental results demonstrate the superiority of our method over state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Accurate cloud and shadow detection is a crucial prerequisite for optical remote sensing image analysis and application. Multi-temporal-based cloud and shadow detection methods are a preferable choice to detect clouds in complex scenes (e.g., thin clouds, broken clouds and clouds with interference from artificial surfaces with high reflectivity). However, such methods commonly require cloud-free reference images, and this may be difficult to achieve in time-series data since clouds are often prevalent and of varying spatial distribution in optical remote sensing images. Furthermore, current multi-temporal-based methods have limited feature extraction capability and rely heavily on prior assumptions. To address these issues, this paper proposes a Siamese network (Siamnet) for cloud and shadow detection based on Time-Series cloudy Images, namely TSI-Siamnet, which consists of two steps: 1) low-rank and sparse component decomposition of time-series cloudy images is conducted to construct a composite reference image to cope with dynamic changes in the cloud distribution in time-series images; 2) an extended Siamnet with optimal difference calculation module (DM) and multi-scale difference features fusion module (MDFM) is constructed to extract reliable disparity features and alleviate semantic information feature dilution during the decoder part. TSI-Siamnet was tested extensively on seven land cover types in the well-known Landsat 8 Biome dataset. Compared to six state-of-the-art methods (including four deep learning-based methods and two classical non-deep learning-based methods), TSI-Siamnet produced the best performance with an overall accuracy of 95.05% and MIoU of 84.37%. In three more challenging experiments, TSI-Siamnet showed enhanced detection of thin and broken clouds and greater anti-interference to highly reflective surfaces. TSI-Siamnet provides a novel strategy to explore comprehensively the valid information in time-series cloudy images and integrate the extracted spectral-spatial–temporal features for reliable cloud and shadow detection. [ABSTRACT FROM AUTHOR]

Feature extraction and matching play integral roles in autonomous vision navigation. As a typical morphological feature, craters can be used as navigation landmarks. However, the spin of small bodies introduces substantial variations in illumination and viewing angles within acquired images, thereby posing challenges to feature extraction and matching algorithms. Addressing these challenges, a deep learning-based approach emerges as pivotal in mitigating the limitations of existing navigation algorithms. This paper presents an algorithm designed to segment noteworthy craters and irregular regions on such small bodies, thereby furnishing crucial autonomous navigation data for missions involving uncharted terrains. By seamlessly integrating deep learning techniques with conventional computer vision-based feature description methods, the algorithm achieves matching segmentation results of noteworthy precision. The experimental results show that the algorithm has better results compared to the existing mainstream segmentation networks. And it can achieve the accurate matching of segmentation results. • Accurate segmentation of small-body craters via deep-learning Algorithms. • Gradient descriptor construction method based on segmentation results. • The proposed method has viewpoint, illumination, scale, and rotation invariance. [ABSTRACT FROM AUTHOR]

Human age and gender prediction from facial images has garnered significant attention due to its importance in various applications. Traditional models struggle with large-scale variations in unfiltered images. Convolutional Neural Networks (CNNs) have emerged as effective tools for facial analysis due to their robust performance. This paper presents a novel CNN approach for robust age and gender classification using unconstrained real-world images. The CNN architecture includes convolution, pooling, and fully connected layers for feature extraction, dimension reduction, and mapping to output classes. Adience and UTKFace datasets were utilized, with the best training and testing accuracies achieved using an 80% training and 20% testing data split. Robust image pre-processing and data augmentation techniques were applied to handle dataset variations. The proposed approach outperformed existing methods, achieving age prediction accuracies of 86.42% and 81.96%, and gender prediction accuracies of 97.65% and 96.32% on the Adience and UTKFace datasets, respectively. [ABSTRACT FROM AUTHOR]

Breast cancer is become the most prevailing and fastest growing disease. In medical imaging, the use of machine learning and deep learning algorithms is essential. Classification of the tumor to predict the chemotherapy response for survival is trivial. In this paper, an innovative Gaussian-Wiener filter combination is used for de-noising the MRI images. These pre-processed images with good image quality are selected for tumor detection. On the basis of these pre-processed outputs, important features are extracted to determine the spatial relationship between the image pixels which results in better texture analysis for the tumor images. Analysis is made on the ISPY-2 trial breast MRI database. Results are analyzed which gives better image quality performance for MRI images. The filters and feature extraction method analyzed is used further in the segmentation and optimization process for breast detection and diagnosis to get the best accuracy of nearly 100%. The results also show better texture analysis for extracting features using GLCM based method. Furthermore, the MRI images for these methods used are explained for better performance in the process of breast cancer detection and diagnosis. [ABSTRACT FROM AUTHOR]

Stress is a psychological or emotional strain that occurs due to adverse experiences in human life. This paper showcases the application of deep learning in detecting stress levels in continuous audio signals in the Distress Analysis Interview Corpus Wizard of Oz (DAIC-WOZ) database. The features that have been experimented with are Gammatone Frequency Cepstral Coefficients (GFCC), Log Filter Bank (Log-Filter Bank), Mel Frequency Cepstral Coefficients (MFCC), chroma, and Linear Predictive Coding (LPC). Five deep learning models were evaluated: Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), Bidirectional LSTM (Bi-LSTM), k-fold CNN with the k value as 5, and a fusion model of CNN, LSTM, and attention. Upon evaluating the performance metrics of all the models, it is concluded that the k-fold CNN model with k as 5 performs well with continuous audio signals. The model has achieved an accuracy of 80% when it is trained on MFCC, GFCC, and Log-F Bank features which are observed to be the optimal features in the stress analysis. [ABSTRACT FROM AUTHOR]

Speech Emotion Recognition (SER) refers to the ability of Machine Learning (ML) and Deep Learning (DL) techniques to accurately predict people's emotional states from speech signals. significant progress has been achieved in the SER domain involving the incorporation of DL models to introduce novel features extraction processes. This paper introduces the use of deep representations learned from the multi-modal Large Language Model (LLM) called ImageBind. These representations were subsequently provided as input to the Nu-Support Vector Machine (Nu-SVM) with RBF kernel for the classification task. The experiments were executed using the IEMOCAP database within the context of a Speaker-Dependent (SD) scenario. The method achieved a noteworthy overall accuracy rate of 80.58% for the four emotions of IEMOCAP, representing a substantial improvement over well-established methods in the existing body of literature. Thus, affirming that the proposed methodology, founded upon ImageBind representations, introduces a novel perspective to the field of SER. [ABSTRACT FROM AUTHOR]

Hydrogen attack is the major failure cause of hydrogen equipment breakdown maintenance. Especially, common problems like cracks frequently occur but are challenging to find while operating, presenting an issue for safety and production. In the process of hydrogen damage evolution, a method for defect state recognition is proposed in this paper. Acoustic emission (AE) technology is used for inspecting the entire hydrogen charging process. The characteristic parameters including the counts and duration of the AE signals are first preprocessed, and the current damage states such as the dislocation propagation, and the occurrence of cracks are identified. Then, a deep learning convolutional neural network is used to create a hydrogen defect recognition (HDR) model with the input of a short-time Fourier transform for the feature vector extraction of various damage status AE signals. Finally, the hydrogen defect recognition experiment revealed that HDR is better in classification accuracy at 98.37% due to dislocation propagation and cracks identification. The study can provide an online damage recognition approach for damage state early warning and evaluation to guarantee hydrogen equipment safety operation. • Monitor the hydrogen damage process using acoustic emission detection technology. • Characteristics of the acoustic sources in hydrogen damage process are studied. • Neural network is used for hydrogen damage signal recognition and diagnosis. • STFT transform image is used as the neural network model input. [ABSTRACT FROM AUTHOR]

The complexity and diversity of texts make it difficult for shallow text classification models to capture deeper text features. Therefore, this paper takes advantage of the BiLSTM-CNN hybrid network based on the self-attention mechanism to extract text features, and constructs a new text feature representation similar to RGB-channel images, that is, tri-channel text feature maps. Drawing on the effectiveness of ResNet and SA-Net in deep networks, we designed a deep feature extraction network to capture deeper features in the text. We use the tri-channel text feature map as the input to the deep feature extraction network, and propose the Deeper Feature Text Classification (DFTC) end-to-end text classification model. Experiments prove that the DFTC model is competitive compared to the most advanced methods on five challenging text classification datasets. • We propose a novel text feature representation in the form of a tri-channel textual feature map. • We designed a deep feature extraction network to capture deeper features in the text. • We construct a deep feature extraction text classification model and achieve impressive results. [ABSTRACT FROM AUTHOR]

Breast cancer diagnosis from mammographic scans is a complex challenge, demanding accurate and efficient methodologies. Existing approaches often rely on time-consuming manual feature extraction or lack scalability when employed with diverse clinical image sets. To address these limitations, this paper presents an innovative solution: the MPC2EL (Multiple Phased Convolution model for Pre-emption of Breast Cancer via Extensive Learning Operations). The core problem tackled by this research is the need for improved breast cancer detection, especially concerning scalability, accuracy, and practicality. To achieve this, MPC2EL leverages extensive learning operations and multiple phased convolutions to process extensive clinical datasets containing both cancerous and benign images. Our novel approach begins with the conversion of these images into multimodal feature sets, encompassing frequency, cosine, wavelet, and Gabor domains. These feature sets enable enhanced classification of images into different cancer stages. The proposed methodology lies in a Dual 1D Convolutional Neural Network (D1D CNN) architecture, capable of categorizing images initially into malignant and benign types and subsequently refining classification for malignant images. Experimental results reveal a substantial 5.9% increase in classification accuracy and a 4.5% improvement in recall compared to conventional deep learning methods. Moreover, the model demonstrates lower processing delay and higher precision, making it a promising candidate for real-world clinical deployment. [ABSTRACT FROM AUTHOR]

• Extracting complex feature information of TIS using multiple encodings. • Capturing hierarchical relationships between features using capsule network. • Improvements are made to the capsule network to enhance performance. • The proposed model improves the prediction performance compared to other models. Genes are the basic units of protein synthesis in organisms, and accurately identifying the translation initiation site (TIS) of genes is crucial for understanding the regulation, transcription, and translation processes of genes. However, the existing models cannot adequately extract the feature information in TIS sequences, and they also inadequately capture the complex hierarchical relationships among features. Therefore, a novel predictor named CapsNet-TIS is proposed in this paper. CapsNet-TIS first fully extracts the TIS sequence information using four encoding methods, including One-hot encoding, physical structure property (PSP) encoding, nucleotide chemical property (NCP) encoding, and nucleotide density (ND) encoding. Next, multi-scale convolutional neural networks are used to perform feature fusion of the encoded features to enhance the comprehensiveness of the feature representation. Finally, the fused features are classified using capsule network as the main network of the classification model to capture the complex hierarchical relationships among the features. Moreover, we improve the capsule network by introducing residual block, channel attention, and BiLSTM to enhance the model's feature extraction and sequence data modeling capabilities. In this paper, the performance of CapsNet-TIS is evaluated using TIS datasets from four species: human, mouse, bovine, and fruit fly, and the effectiveness of each part is demonstrated by performing ablation experiments. By comparing the experimental results with models proposed by other researchers, the results demonstrate the superior performance of CapsNet-TIS. [ABSTRACT FROM AUTHOR]

This paper presents a novel Deep Learning (DL) framework for cotton plant disease prediction based on Explainable Artificial Intelligence (XAI) and Copula entropy based-Grey Wolf Optimization (GWO) algorithm. The suggested framework uses a cotton plant image as input and pre-processes it to produce an image with enhanced contrast. Features are extracted from the leaf images with the ResNet50 CNN model. A crucial pre-processing model known as feature selection helps to improve the effectiveness of image classification by deleting extraneous or irrelevant features. Therefore, the Gray Wolf optimization (GWO) algorithm which is a global search method with potential use in feature selection is employed in this paper. The proposed framework introduces Copula entropy (CE) as an indicator of association to create the GWO's initial population and enhance the GWO feature engineering process. For the GWO initialization procedure, CE has been used to choose the most significant features which substantially improved the quality of the GWO starting population and as a result improved the performance of the proposed CE-based GWO algorithm by 78.57 % faster than the traditional GWO as stated by the time complexity analysis. In addition, Feature importance explanation is determined using XAI layer. The final classification is achieved using Random Forests (RF) classifier which is an ensemble learning approach. According to the experimental results, the suggested model has a classification accuracy of 99 % and a mean squared error of 0.0383. Furthermore, the proposed model has been compared to state-of-the-art algorithms and the results showed that the proposed model has the superiority performance. The proposed model can therefore be used to track a variety of cotton areas to enable faster analysis and response, resulting in higher productivity. [Display omitted] • Hybrid DL-GWO model is proposed to detect cotton diseases. • Copula Entropy is proposed to enhance the GWO initialization. • XAI is used as a second layer for explaining and enhancing the feature importance. • The proposed model achieved promising accuracy when compared to related models. [ABSTRACT FROM AUTHOR]

Data-driven soft sensors have emerged as indispensable tools for predicting quality variables in complex industrial processes because of their cost-effectiveness and ease of maintenance. In particular, soft sensors based on deep learning have been utilized in extensive research and successful applications in recent years. However, traditional deep learning methods capture hierarchical data features by minimizing global fitting errors, neglecting the local structural characteristics implied in the original data. In this paper, we propose a new deep learning method for soft sensor development. Utilizing autoencoders as the foundational architecture of our network, a new semisupervised strategy is adopted for layerwise pretraining optimization. On the one hand, more representative data features are extracted by maintaining the local spatiotemporal structure of the data; on the other hand, layer-by-layer supervised learning is employed to identify the critical features that are aligned with the ultimate task, which aids in obtaining the optimal network parameters and improving the resulting prediction accuracy. Subsequently, a local spatiotemporal structure-preserving stacked semisupervised autoencoder (LSP-SuAE) is established. To evaluate the feasibility and effectiveness of the proposed approach, experiments are carried out in a real industrial process. A soft sensor based on the LSP-SuAE is developed to predict the rate of ethylbenzene conversion during the dehydrogenation of styrene. The experimental results demonstrate that, compared to five other common or similar data-driven modeling methods, LSP-SuAE exhibits higher prediction accuracy and better stability. • A deep learning method for soft sensor modeling uses a semisupervised pretraining strategy integrating spatiotemporal structure preservation. • The developed soft sensor is deployed in an industrial scenario where soft sensing technology has not been widely adopted. • This paper provides a theoretical and practical foundation for deep learning to address problems involving spatiotemporal data structures. [ABSTRACT FROM AUTHOR]

The emergence of artificial intelligence (AI) and its related capabilities has led industries to rethink the existing practices of conventional supply chain management and data analysis. Machine learning (ML), Deep Learning (DL) and their unique ability to predict future data and classify data have led to important research in the supply chain (SC) domain, particularly in identifying and prioritizing supply chain risks. This paper proposes several DL methodologies to exploit the benefit of DL, particularly to identify whether any product will be delivered late due to any unforeseen reason in a complex SC system. Four different DL architectures (Simple-LSTM, Deep-LSTM, 1D-CNN, and TCN-1DSPCNN models) are proposed to extract features, while six variant classifiers: Softmax, random trees (RT), random forest (RF), K-nearest neighbor (KNN), artificial neural network (ANN), and support vector machine (SVM), were used to classify delay or non-delay information. By seamlessly capturing intricate temporal dependencies, these DL models enhance accuracy in robustly identifying supply chain late orders. Leveraging their hierarchical feature learning, these proposed DL models excel in recognizing subtle patterns and correlations, making them ideal for classifying late orders within the supply chain. Their parallel processing prowess facilitates real-time decision support, allowing organizations to address potential delays and allocate resources effectively and proactively. Five-fold cross-validation is presented to avoid over-fitting and to prove the efficiency of the proposed DL models. The total accuracies of the six ML classifiers are 74.03, 75.81, 93.35, 87.72, 93.59, and 95.10, respectively, while the maximum accuracies obtained from four proposed DL methodologies obtained an accuracy of 97.6, 98.63, 100, 100% respectively using the SVM classifier for predicting late orders based on five-fold cross-validation. • This paper investigates a few DL approaches to extract features of SC data. • Both RNN and CNN are applied in the same model. • An improved CNN model has been proposed for feature extraction. • An online dataset is employed to validate the proposed DL architectures. [ABSTRACT FROM AUTHOR]

With the popularization of computer technology, the number of malware has increased dramatically in recent years. Some malware can threaten the network security of users by downloading and installing, and even spreading widely on the Internet, causing consequences such as private data leakage in the operating system, extortion, and network paralysis. In order to deal with these threats, researchers analyze malicious samples through various analysis techniques, which are usually divided into static and dynamic analysis based on the principle of whether the code needs to be executed or not. This paper analyzes in detail several classical methods of feature extraction in malware detection techniques. With the technological development of artificial intelligence, deep learning is gradually being introduced into malware detection, which does not require the identification of professional security personnel and greatly improves the generalization ability of detection. In the paper, text-based detection methods, image visualization-based detection, and graph structure-based detection techniques are reviewed according to different feature extraction methods. In addition, the paper compares 26 datasets that have been commonly used in recent years applied in the research field and explains the main contents and specifications of the datasets. Finally, a summary and outlook of the malware research field is given. [ABSTRACT FROM AUTHOR]

• An unsupervised feature extraction method of Li-ion batteries' EIS is proposed. • The latent variables are extracted automatically without prior knowledge. • The extracted latent variables are related to battery aging. • The capacity of batteries for uneven usage during degradation is estimated. • The proposed method generates reliable EIS data. In this paper, we propose an unsupervised feature extraction technique using a variational autoencoder and generative adversarial network (VAEGAN) that automatically extracts meaningful latent variables from the electrochemical impedance spectra (EIS) of lithium-ion (Li-ion) batteries. These variables accurately reflect the degree of Li-ion battery degradation and are closely related to its capacity. By analyzing the latent variables, it was found that the network can learn the effect of Li-ion battery capacity degradation on the EIS. The extracted latent variables are then used as input features for Gaussian Process Regression (GPR) to estimate the capacity of Li-ion batteries under uneven usage and with unknown historical data. The results show that the capacity prediction method proposed in this paper can significantly reduce prediction errors compared to the current state-of-the-art methods. The method not only provides a more reliable capacity estimation but is also robust to highly noisy EIS data, making it more suitable for practical application scenarios. [Display omitted] [ABSTRACT FROM AUTHOR]

Many deep neural network-based methods have recently been proposed for object detection due to the significant success of deep learning in computer vision. However, existing object detection methods typically extract the appearance features of objects from single image so that they usually suffer from poor performance in detecting micro Unmanned Aerial Vehicle (UAV), because micro UAV lacks of rich color, shape and texture information. To address this issue, we introduce the temporal information of objects from videos and develop a Spatio-Temporal Attention Module (STAM) to efficiently enhance feature map extraction for detecting micro UAV, and then integrate STAM into YOLOX to develop a video object detector for micro UAV. Meanwhile, we propose a lightweight Spatial Pyramid Pooling (SPP) module termed Group Simplified Spatial Pyramid Pooling-Fast with Cross Stage Partial (Group SimSPPFCSP) for the backbone's final stage layer to efficiently and lightly extract more semantic information, and we propose a neck with rich propagation pathways (NRPP) to facilitate the effective propagation of spatial and temporal information across different levels. Furthermore, we propose two data augmentation operations including SeqMosaic and SeqMixUp, to augment video data for video object detection. Experimental results show that our model can achieve competitive precision (with 5.0 mAP and 8.1 mAP S m a l l improvement) while maintaining real-time inference speed (35.3 fps). • This paper develops a spatio-temporal attention module which integrates spatial and temporal context information to enhance feature extraction of micro UAV. • This paper presents a lightweight SPP module named Group SimSPPFCSP to efficiently extract more semantic information and a neck with rich propagation pathways to facilitate the effective propagation of spatial and temporal information across different levels. Finally, the detection head is further optimized to focus on detecting small objects. • This paper proposes two data augmentation operations to improve the video detector for small objects. [ABSTRACT FROM AUTHOR]

• Paper proposes Deep and Wide Nonnegative Matrix Factorization with embedded regularization (DWNMF) as an end-to-end model. • It prevents overfitting via embedded regularization, and decreases vanishing gradient via training layers independently. • Model size can grow incrementally to achieve the performance. It saves memory since it only holds each layer's parameters. • Model size can grow incrementally to achieve the performance. It saves memory since it only holds each layer's parameters. • Experimental results showed that DWNMF performs better than end-to-end feature learning models in complexity and CPU time. End-to-end learning is an advanced framework in deep learning. It combines feature extraction, followed by pattern recognition (classification, clustering, etc.) in a unified learning structure. However, these deep networks face several challenges such as overfitting, vanishing gradient, computational complexity, information loss in layers, and weak robustness to noisy data/features. To address these challenges, this paper presents Deep and Wide Nonnegative Matrix Factorization (DWNMF) with embedded regularization for the feature extraction stage of the end-to-end models. DWNMF aims to identify more robust features while preventing overfitting via embedding regularization. For this purpose, DWNMF integrates input data with its noisy versions as diverse augmented channels. Then, the features in all channels are extracted in parallel using distinct network branches. The parameters of this model learn the intrinsic hierarchical features in the channels of complex data objects. Finally, the extracted features in different channels are aggregated in a single feature space to perform the classification task. To embed regularization in the DWNMF model, some NMF neurons in the layers are substituted by random neurons to increase the stability and robustness of the extracted features. Experimental results confirm that the DWNMF model extracts more robust features, prevents overfitting, and achieves better classification accuracy compared to state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Metallic fuels, particularly U 10Zr, are promising candidates for next-generation sodium-cooled fast reactors. Irradiation of nuclear fuels in reactors can lead to the formation of solid and gas fission product which subsequently forms microstructural pores, deteriorating fuel performance. Due to the massive amount of pores and complex phases formed, a quantitative description of fission gas pores is not yet available, preventing the development of microstructure-informed fuel performance modeling for fuel qualification. This paper applied a pre-trained deep learning model to ∼10,260 high magnification scanning electron microscopy images. This method increased the accuracy of fission gas pore segmentation and allows statistical features to be extracted which cannot be achieved manually. A pre-trained decision tree model worked on the segemenation resutls and further classified the pores into different categories to produce a correlation between the pores, movement of lanthanides, and temperature gradient during irradiation. This paper emphasizes the potentials of machine learning models to accelerate fuel research, development, and qualification for advanced reactors. [Display omitted] • Deep learning model is used to detect pores in ∼10,260 scanning electron microscopy images. • About 230,000 pores are detected and classified based on their impact on nuclear fuel performance. • Pore statistics and orientations help to understand fuel performance from post-irradiation examination. [ABSTRACT FROM AUTHOR]

This paper presents the novel task of bluespotted ribbontail (BR) ray (Taeniura lymma) recognition using deep learning based computer vision methods to enable the identification of specific individuals of this species. Mapping the specific individuals in relation to location and time will allow marine researchers to understand their movement patterns, habitat choice, life span, size of the population and more – data which could allow monitoring and establishing a tailor-made conservation plan for this species. Our work is pioneer on this recognition problem. We give a detailed description of the three basic steps of detection, feature extraction and recognition in this vision problem and perform experiments to explore the system configuration and what improves the performance. A feature extraction enhancement as well as a crucial effect of a split into different main poses are demonstrated. Though the precision results achieved in this paper are still moderate and should be further improved, they are nevertheless promising and reasonable for practical use if the six best matches are chosen. For this scenario, almost 85% precision for upper-pose model, and almost 80% precision for left- and right-pose models, are achieved demonstrating the feasibility of the pipeline suggested as well as opportunities for improvement. • We present a pioneer novel application of deep learning based computer vision methods for bluespotted ribbontail (BR) ray recognition, to facilitate the ecological study of this species. • We explored what improves the performance. • An effective feature extraction enhancement technique is applied in our framework. • The crucial effect of a split into different main poses in order to achieve BR ray recognition is demonstrated. [ABSTRACT FROM AUTHOR]

• The paper introduces an adaptive dynamic graph for multi-label classification of ECG signals. The dynamic graph is generated through adaptive capturing of the category relationships in the signals, thereby improving the correlations among different labels. • The paper proposes the ED-DGCN, which is mainly composed of three modules. The encoder block is utilized to extract signal feature information. The decoder block can define semantic regions and generate a categorical representation of content perception for each ECG signal. The DGCN mainly models category representation relationship learning. • The transfer learning method is adopted to pre-train the network model on a large dataset and to load the training weight onto the CPSC2018 dataset. The effectiveness of the proposed method for arrhythmia classification was verified by experiments on the CPSC2018 dataset. Cardiac arrhythmia is a condition characterized by abnormal heart electrical activity, posing a significant health risk to patients. Despite significant progress in existing arrhythmia classification methods, the accuracy of network-based classification still requires improvement. In the paper, we propose an effective classification algorithm called Encoder-Decoder Dynamic Graph Convolutional Network (ED-DGCN). In the encoder stage, a convolutional neural network(CNN) is utilized to extract feature information from electrocardiogram signals. The decoder stage employs a bidirectional long short-term memory network to define semantic regions and generate corresponding class relationships. A dynamic graph convolutional network module is constructed to capture label dependencies by obtaining a dynamic adjacency matrix. Additionally, transfer learning is employed by pre-training the proposed network on a large dataset to further enhance convergence performance and classification accuracy. The proposed method is verified on the China Physiological Signal Challenge 2018 (CPSC2018) dataset. Experimental results demonstrate that the average F1-score of our method achieved 0.843, which is 1.6 % improved over existing methods. The proposed method effectively enhances the classification performance of the network and provides an effective solution for multi-label classification tasks in cardiac arrhythmias. The method holds promising prospects for clinical applications. [ABSTRACT FROM AUTHOR]

• The classification of brain tumors (BTs) is an expensive intricate challenge in the section of clinical image analysis. Radiologists were able to reliably detect tumors using machine learning (ML) algorithms without the need for extensive surgery. • The efficacy of this paper is examined using classification accuracy. • The suggested model demonstrated its advantage over the currently used approaches for classifying BTs from MR images. The classification of brain tumors (BTs) is an expensive intricate challenge in the section of clinical image analysis. Radiologists were able to reliably detect tumors using machine learning (ML) algorithms without the need for extensive surgery. Hand-crafted features are necessary for the conventional ML classifiers, which take a lot of effort. Deep Learning (DL), on the other hand, has gained popularity recently for identification and categorization-based functionalities because of its effective feature extraction capabilities. Creating the most precise DL architecture for classifying BTs is a huge task. These challenges inspired us to develop a cutting-edge and highly accurate framework based on DL and computational methods that evolve to effectively create the hybrid framework automatically for categorizing three distinct types of brain cancers on a vast library of MR images. We present a unique EBT Deep Net DL model that uses a substantial convolutional neural network (CNN) technique for the classification of three distinct forms of brain cancers: gliomas, tumors in brain meningeal tissue, and tumors in pituitary glands. Pretrained deep CNN models like ResNet50V2, ResNet152, DenseNet201, MobileNetV3, and VGG19 were used to extract deep features. The efficacy of this paper is examined using classification accuracy. In comparison to the most recent classification findings from ResNet50V2, ResNet152, DenseNet201, MobileNetV3, and VGG19, the proposed methodology had the greatest accuracy. The models' respective accuracy on a classification test is as follows: VGG19 scored 97.46 %, ResNet152 scored 98 %, DenseNet201 scored 95.44 %, ResNet50V2 scored 90.04 %, MobileNetV3 scored 97.15 %, and the ensemble model outperformed all of them with an accuracy of 98.8 %. The suggested model demonstrated its advantage over the currently used approaches for classifying BTs from MR images. [ABSTRACT FROM AUTHOR]

Deep learning is widely used in Bearing Fault Diagnosis (BFD). Nonetheless, practical industrial production often generates a large amount of industrial noise. These noises exhibit randomness and complexity, which puts forward higher requirements for diagnosis algorithms. Certain studies have tackled the issue of anti-interference in high-noise environments (S N R ≤ 0 dB) by increasing the complexity of the model. However, due to the excessive number of parameters and computation, such models cannot be deployed on low-end edge devices. Balancing resource consumption and accuracy has become a major challenge in BFD modeling research. To solve the above problems, this paper proposes a new Transformer architecture model called LTFAFormer. The LTFAFormer is capable of achieving high-precision diagnostics on low-end edge devices and shows greater noise resistance. In terms of processing sequence information, Transformer has proven to be superior to other solutions. However, when dealing with longer sensor signal data containing complex noise, the traditional self-attention mechanism not only cannot effectively extract fault features, but also generates more computational complexity than CNN. To address this issue, we propose a novel time–frequency dual-channel parallel attention mechanism. Our approach enhances the feature extraction capability of the model by expanding the attention computation scale and reduces the computational resource consumption of the model by optimizing the model structure. To validate the effectiveness of LTFAFormer, we present two cases to demonstrate that LTFAFormer has higher prediction accuracy while satisfying lightweight. Especially in high-noise environments, LTFAFormer has stronger robustness. In this paper provides a new set of feasible strategies for the practical deployment of BFD models in practical industrial environments. • The LTFAFormer, a lightweight and noise-robust model, is ideal for low-end devices. • Added a Dual-Channel Attention Module to expand traditional self-attention scale. • Developed a Cross-fusion mechanism for attention to enhance feature fusion. • Optimized the model's embedding module based on sensor signal traits. • Our model outperforms others in bearing fault diagnosis and is robust in noise. [Display omitted] [ABSTRACT FROM AUTHOR]

• The article proposed a novel methodology for automated surface electromyography (sEMG) signal analysis focusing on diverse machine learning models, and hyperparameter optimization achieving an F1-Score of 98.71 % with the XGBoost algorithm. • Hyperparameter tuning was performed for each model using the scikit-learn library. The paper provides detailed information on the best hyperparameter settings for XGBoost, Extra Trees, LGBM, SVM, and Neural Networks. • The models were trained on a dataset featuring muscle contractions and extensions in distinct body regions. This diverse training set aimed to capture the complexities and variations encountered in real-world surface electromyography (sEMG) applications. • The paper employs non-parametric tests, including Friedman's test and post-hoc tests, to assess the significance of differences among the models. Visualizations and statistical analysis provide insights into algorithm performance based on F1-Score, Precision, and Recall. • The models showcased practicality for real-world applications, supporting clinicians in diagnosing neuromuscular disorders and advancing human–machine interfaces. This paper presents a methodology for automatically detecting muscular activity by denoising, extracting features, and classifying surface electromyography (sEMG) signals. The proposed methodology utilizes the Discrete Wavelet Transform (DWT) and Willison's Amplitude Algorithm (WAMP) for feature extraction. Five classification methods, including Neural Networks (NN), Classification Vector, XGBoost, Light Gradient Boosting Machine (LGBM), and ExtraTree, were evaluated using F-Measure, Precision, and Recall as performance metrics. Through k-fold cross-validation, the XGBoost algorithm, when combined with the Eigen values feature, achieved the highest training performance with an F1-Score of 98.71 %. For the test group, the LGBM classifier using WAMP, and NN with both WAMP and Eigen values as features, demonstrated the best average performance with F1-Scores of 96.52 ± 3.45 % and 96.52 ± 3.07 %, respectively. These results highlight the precision and performance of the proposed approach in detecting EMG signals. Moreover, the framework has the potential to support clinicians in diagnosing neuromuscular disorders and developing human–machine interfaces. [ABSTRACT FROM AUTHOR]

The reliability of bolt connections significantly impacts the operational state and lifespan of industrial equipment. Vision-based noncontact methods exhibit high efficiency in bolt loosening detection. However, limited image features hinder measurement accuracy. To improve bolt loosening detection performance, this paper proposes a novel deep learning backbone, the high-resolution cross-scale transformer, to extract high precision keypoints for bolt three-dimensional model construction. Simultaneously, a monocular vision measurement model is established to get the bolt exposed length and evaluate the connection loosening state. The proposed backbone hybridizes the advantages of high-resolution architecture and transformer, realizing global information aggregation and fine-grained image details. A simplified module, dual-scale multi-head self-attention, is designed to reduce the computational redundancy caused by the implementation of high-resolution multi-branch architecture. In the experiment section, the high-resolution cross-scale transformer outperforms other keypoint detection baselines, achieving the top one performance with 91.6 average precision and 84.9 average recall. The monocular vision measurement model realizes a 0.053 mm error with a 0.028 mm standard deviation, satisfying the industrial implementation requirement. Additionally, the model is tested on different industrial situations and an additional outside dataset, indicating the model's robustness and actual environment adaptability. • A monocular vision measurement method for bolt loosening detection is proposed. • It is the first attempt to introduce the transformer mechanism in bolt keypoint detection. • A new keypoint detection backbone is proposed in this paper for bolt feature extraction. • The 3D exposed length of bolts is calculated with a monocular vision system. • The bolt loosening detection method is validated. [ABSTRACT FROM AUTHOR]

State-of-Health (SOH) estimation of lithium-ion batteries is critical for reliable and efficient energy storage and usage. Traditional SOH estimation approaches either compromise on real-time performance or on estimation accuracy. This paper introduces an end-cloud collaboration approach that synergistically combines a cloud-based deep learning model with an end-based empirical model to achieve both high-accuracy and real-time SOH estimation. Utilizing the Transformer architecture, the cloud-side model employs multiple features extracted through signal analysis technology. On the other hand, the end-side employs a fast, iterative double exponential empirical model that provides real-time estimates. Two methods of end-cloud collaborative updating are presented, both leveraging Kalman Filter and Unscented Kalman Filter (UKF) to integrate and iteratively update the models. Evaluation shows that the cloud-side model achieves a Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) of around 0.8 %, while the end-cloud collaboration system maintains these metrics at 1 %, successfully capturing nonlinear capacity rebound phenomena. The work contributes to the domain by offering an approach that balances computational load and latency, making it suitable for online applications. Under the Cyber Hierarchy and Interactional Network (CHAIN) framework, this end-cloud collaboration architecture holds promise for high-accuracy, real-time health management of lithium-ion batteries across their full lifecycle. The entire end-cloud collaboration system includes a cloud based deep learning model, an end-side fast empirical model and an end-cloud collaboration algorithm. On the cloud-side, firstly, data preprocessing such as data cleaning and filtering is performed on the data. Then, multiple features are extracted based on various signal analysis methods, including differential thermal voltammetry analysis, incremental capacity analysis and equal voltage discharge interval discharge time analysis. The deep learning model based on Transformer is constructed. On the end-side, the fast model is constructed based on a dual exponential empirical model. And the end-cloud collaboration algorithm is arranged on the end-side. Two end-cloud collaboration methods are used simultaneously, including result fusion and iterative update based on Kalman filter, and online parameter identification based on UKF by sequentially linking end and cloud models. Finally, under the cyber hierarchy and interactional network framework, the end-cloud collaboration approach proposed in this paper has the potential to realize high-accuracy and high-real-time health management of lithium batteries throughout the full life cycle. [Display omitted] • A high accuracy and real-time SOH estimation method is presented. • Deep learning model on cloud-side is constructed with Transformer model as the core. • Feature extraction is based on multi feature collaborative analysis. • Two end-cloud collaboration methods are proposed to update the end-side. [ABSTRACT FROM AUTHOR]

Over the past decades, deep learning techniques have attracted increased attention from various research and industrial domains aligned with the development of Industry Internet-of-Things(IIoT). Specifically, with the advantage of data-driven methods, industrial organizations are seeking novel proactive strategies supported by analytic models to guarantee the quality of their production by observing degradation or predicting failure ahead of the occurrence of the component or asset. Predictive strategies are expected to promise the influence of unnecessary maintenance interruptions and mitigate the consequence of that, hence, extending the remaining useful life of products. This paper conducts a survey of the utilization of deep learning technologies on engineering applications where they provide satisfactory solutions with respect to specific data types or input signals. 106 primary papers are reviewed on deep learning–driven approaches which mainly explore five of the most popular architectures in the application of predictive maintenance. The main content of this paper summarizes the common advantages of each architecture and, accordingly, points out their limitations, as well as describes the application scopes of fully connected deep neural networks, convolutional neural networks, stacked autoencoders, deep belief networks, and deep recurrent neural networks. Based on the technique discussion for each of them, we intend to provide a comprehensive understanding and guidance of the appropriate usage of deep learning architectures to devise an effective predictive maintenance strategy for the scientific and industrial developers whose expertise lies in the prior domain knowledge of multi-source isomerization data. Moreover, the main content demonstrated the summarization of the decisive factor by which the incremental stages of the approaches were determined, fundamentally including the dataset specification, feature extraction, and the integration of deep learning approaches. [ABSTRACT FROM AUTHOR]

Human activity recognition (HAR) is a crucial detection technique extensively employed in various contexts demanding accurate identification of human actions. Currently, mainstream HAR approaches frequently rely on sequential data generated by wearable sensors, with a primary focus on extracting feature representations that align with various human activities. Nevertheless, complex and diverse human activities encompass various kinds of features and substantial variations in the features emphasized across different activities, making it challenging for current artificial intelligence methods to achieve accurate recognition. To address this challenge, this paper introduces a network framework called the Multi-Feature Combining Attention Neural Network (MFCANN). This framework incorporates diverse feature extraction, along with both local and global feature attention mechanisms. It is composed of stacked Multi-Feature Combining Attention Blocks (MFCAB) designed by us. In contrast to traditional convolutional methods, MFCAB parallelly stacks various convolutional components, enabling the extraction of a broader range of features from human activity data. Additionally, we propose the Intra-Module Attention Block (Intra-MAB) and the Inter-Module Attention Block (Inter-MAB), which simultaneously focus on local fine-grained features within module feature maps and global distinguishing features across module feature maps. This aims to achieve more targeted feature learning, reinforcing the network's ability to distinguish between different human activities. Abundant experimental results demonstrate that the proposed MFCANN in this paper outperforms current mainstream deep learning algorithms in HAR tasks, achieving recognition accuracies of 0.9813, 0.9324, and 0.9930 on the UCI-HAR, USC-HAD, and RealWorld datasets respectively. We release our code at https://github.com/RangerHeart/MFCANN. [ABSTRACT FROM AUTHOR]

• The study introduces MEEL (Mel-Frequency Energy Line) features as a new set of acoustic properties for classifying diseased voices. MEEL characteristics encompass the spectral attributes of abnormal voices, offering a distinctive and efficient representation for discerning various vocal disorders. • The paper outlines the creation and execution of an innovative classification model called SVM-TabNet. This approach synergistically integrates the robust capabilities of support vector machines (SVM) with TabNet, an attention-based decision-making mechanism. The combination of these two methodologies improves the accuracy of classification, especially when analyzing diseased voices. • The research showcases the efficacy of the SVM-TabNet model in enhancing the precision of pathological voice classification. Using MEEL features, the Model demonstrates exceptional efficacy in differentiating between normal and abnormal vocal states compared to current methodologies. • The paper's notable contribution is its assessment of the SVM-TabNet model across various datasets encompassing different types of pathological voice problems. The proposed approach is designed to apply to various voice disorders, ensuring its potential for use in varied clinical scenarios. This makes it a versatile solution with high generalizability and resilience. • The research paper thoroughly evaluates several techniques used to classify diseased voices. Researchers conducted extensive testing and assessments to demonstrate the superior performance of the SVM-TabNet model compared to other models in terms of accuracy, sensitivity, and specificity. This shows its potential applicability in real-world healthcare environments. • The paper concludes by providing valuable insights and suggesting recommendations for future research in pathological speech analysis. Possible avenues for further research in this crucial field of medical diagnostics involve examining supplementary acoustic characteristics, enhancing the Model's structure, or studying the incorporation of sophisticated machine-learning methods. These efforts would contribute to a roadmap for future progress. In clinical settings, early diagnosis and objective assessment depend on the detection of voice pathology. To classify anomalous voices, this work uses an approach that combines the SVM-TabNet fusion model with MEEL (Mel-Frequency Energy Line) features. Further, the dataset consists of 1037 speech files, including recordings from people with laryngocele and Vox senilis as well as from healthy persons. Additionally, the main goal is to create an efficient classification model that can differentiate between normal and abnormal voice patterns. Modern techniques frequently lack the accuracy required for a precise diagnosis, which highlights the need for novel strategies. The suggested approach uses an SVM-TabNet fusion model for classification after feature extraction using MEEL characteristics. MEEL features provide extensive information for categorization by capturing complex patterns in audio transmissions. Moreover, by combining the advantages of SVM and TabNet models, classification performance is improved. Moreover, testing the model on test data yields remarkable results: 99.7 % accuracy, 0.992 F1 score, 0.996 precision, and 0.995 recall. Additional testing on additional datasets reliably validates outstanding performance, with 99.4 % accuracy, 0.99 F1 score, 0.998 precision, and 0.989 % recall. Furthermore, using the Saarbruecken Voice Database (SVD), the suggested methodology achieves an impressive accuracy of 99.97 %, demonstrating its durability and generalizability across many datasets. Overall, this work shows how the SVM-TabNet fusion model with MEEL characteristics may be used to accurately and consistently classify diseased voices, providing encouraging opportunities for clinical diagnosis and therapy tracking. [ABSTRACT FROM AUTHOR]

• Proposed FSDnet has good performance in crop plant phenotype segmentation. • Introduced Gaussian density can reduce the calculation and improve accuracy. • By embedding FSDnet, the traditional models can work well in different densities. • We construction a new field 3D deep learning dataset including strawberry and apple. The accurate segmentation of fruit phenotypes in the field is of great significance for agricultural automation in the 3D scene. Although the existing fruit segmentation based on 3D point cloud has made great progress, in the complex field environment, due to lighting, leaf occlusion, shooting angle and other problems, the point cloud obtained by depth camera often has the problem of multiple voids and discrete points, which seriously affects the accurate segmentation of fruit phenotype. This paper proposes a embedding subnetwork FSDnet based on density-based feature extraction and feature propagation and embeds it in the novel segmentation networks, which effectively improves the segmentation accuracy of the point cloud phenotype in multi-hole and multi-discrete fruits, including (1) The density-based point cloud feature extraction and feature propagation theory is proposed to alleviate the problem of perception degradation in fruit edge point caused by discrete points and holes caused by imcomplete point cloud in the agriculture scene. (2) A density-adaptive embedding semantic segmentation framework FSDnet is proposed, and embedding the classical point cloud neural network can significantly improve the segmentation accuracy of the fruit phenotypes with multiple holes and discrete points in the traditional network. (3) This paper made a strawberry dataset and tested the designed new neural network on both strawberry and apple filed dataset. After FSDnet is embedded on different novel net, almost all net have been improved. We verified the performance of FSDnet in different density states in agricultural scenarios, mitigated the negative impact of density on segmentation accuracy, proving that it can adapt to different point cloud density in agricultural scenarios in comparison between Gaussian density and other two traditional density schemes, Gaussian density reduces the computational traffic (0.58G) of the network while maintaining similar performance to the other two densities, proving the superiority of assuming a Gaussian density. [ABSTRACT FROM AUTHOR]

Object detection, as one of the most fundamental and essential tasks in the field of computer vision, has been the focus of unremitting efforts by researchers, who are committed to modifying the neural network structure in order to improve the accuracy of object detection and expedite task execution. As the application scope continues to expand, small object detection has gradually emerged as a crucial branch in the field of object detection. In this paper, the development history of object detection algorithms is introduced, the concept of small objects is introduced, and the current problems and challenges faced by small object detection are outlined. In this paper, the network structure is disassembled from a macroscopic point of view, and improved algorithms such as enhanced data augmentation, improved feature extraction, superior feature fusion, and refined loss functions are described in detail. Furthermore, the paper explores a series of emerging and improved algorithms for small object detection. It encompasses the introduction of advanced strategies such as unsupervised learning, end-to-end training, density cropping, transfer learning, and anchor-free approaches. The paper provides a comprehensive list of commonly used general-purpose datasets and domain-specific datasets for small object detection tasks, offering performance comparisons of the mentioned improved algorithms. In conclusion, the paper summarizes and provides an outlook on current small object detection algorithms, furnishing the reader with a thorough understanding of the field and insights into future directions. • GAN can improve the resolution of images containing small objects. • The receptive field can be expanded by dilated convolution. • Multi-scale feature extraction is often used in small object detection. • Small object datasets are mostly concentrated in remote sensing. [ABSTRACT FROM AUTHOR]

Schizophrenia (SCZ) is a severe mental and debilitating neuropsychiatric disorder that disrupts a person's thought processes, emotions, and behavior. Due to misdiagnosis, self-denial, and social stigma, many SCZ cases go untreated. Magnetic resonance imaging (MRI) is an excellent noninvasive tool for soft tissue contrast imaging because it provides crucial data on tissue structure size, position, and shape. The Resnet50 network is a deep residual learning framework used for feature extraction. Random-vector functional link network (RVFL) is an example of a single-hidden-layer feedforward network in which input features and hidden layer features are fed to the output layer. In this paper, we introduced a kernel ridge regression-based random vector functional link (KRR-RVFL) classifier which focuses on addressing the linearity issues in RVFL by designating the kernel function in the input layer for the precise diagnosis of SCZ. The genetic algorithm (GA) seeks to minimize the loss function by optimizing the weights and biases of the KRR-RVFL network. The classification performance is investigated on the SCZ and cognitive normal (CN) subjects, collected from the available open neuro platform, including 99 participants. The results of the suggested network show superior performance to the recent state-of-the-art networks in terms of accuracy 93.66%, sensitivity 92.22%, specificity 95.17%, precision 95.33%, F-measure 93.74%, and G-mean 93.68%. The performance metrics demonstrated the applicability of this framework for assisting clinicians in the automatic, precise evaluation of SCZ. • This paper proposes Schizophrenia diagnosis model using KRR-Opt RVFL classifier. • ResNet50 is deployed for the extraction of complex features from pre-processed data. • GA is incorporated with KRR-RVFL to optimize the parameters of the classifier. • The performance of the proposed KRR-Opt RVFL is compared with the state-of-the-art classifiers. [ABSTRACT FROM AUTHOR]

Analyzing functional brain networks (FBN) with deep learning has demonstrated great potential for brain disorder diagnosis. The conventional construction of FBN is typically conducted at a single scale with a predefined brain region atlas. However, numerous studies have identified that the structure and function of the brain are hierarchically organized in nature. This urges the need of representing FBN in a hierarchical manner for more effective analysis of the complementary diagnostic insights at different scales. To this end, this paper proposes to build hierarchical FBNs adaptively within the Transformer framework. Specifically, a sparse attention-based node-merging module is designed to work alongside the conventional network feature extraction modules in each layer. The proposed module generates coarser nodes for further FBN construction and analysis by combining fine-grained nodes. By stacking multiple such layers, a hierarchical representation of FBN can be adaptively learned in an end-to-end manner. The hierarchical structure can not only integrate the complementary information from multiscale FBN for joint analysis, but also reduce the model complexity due to decreasing node sizes. Moreover, this paper argues that the nodes defined by the existing atlases are not necessarily the optimal starting level to build FBN hierarchy and exploring finer nodes may further enrich the FBN representation. In this regard, each predefined node in an atlas is split into multiple sub-nodes, overcoming the scale limitation of the existing atlases. Extensive experiments conducted on various data sets consistently demonstrate the superior performance of the proposed method over the competing methods. • Adaptive and integrated construction and analysis of hierarchical FBN is proposed. • It overcomes the scale limitations of the traditional predefined brain atlases. • The superiority has been consistently demonstrated in MCI classification tasks. [ABSTRACT FROM AUTHOR]

Quick response codes are widely utilized in commodity anti-counterfeiting, and their traceability via smartphones is an effective authentication method. However, quick response codes are susceptible to illegal duplication, and the images captured by smartphones are easily affected by changes in light and environmental noise, leading to unsatisfactory verification results. To address these issues, this paper introduces a novel approach by combining the squeeze-excitation attention module with bottleneck residual block. It presents a squeeze-excitation bottleneck residual network for printer source identification of quick response codes. The squeeze-excitation attention module pays more attention to the features that represent the printer attributes, reduces the interference of useless information, and has low computational consumption; while the bottleneck residual block has the advantages of few parameters, strong feature extraction capability, and good expandability. Thus, the performance can be improved effectively only with a small increasing in parameters. The experimental results verify that the proposed method achieves an accuracy of 98.77% under smartphone capture conditions, it outperforms other convolutional neural network-based methods in terms of identification accuracy. The deep learning model proposed in this paper can be generalized and applied to the printer source identification of paper content in the civil, criminal investigation and judicial fields. [ABSTRACT FROM AUTHOR]

• Improving the quality of Federal Learning encrypted alignment data. • Use Gaussian mixture clustering to cluster samples and set a threshold to filter samples. • Use the encrypted sample attribute searching algorithm to fill in the missing value of the sample. • Design the combination model of variation auto-encoder Gaussian hybrid clustering and federated learning. Federated learning can achieve multi-party data-collaborative applications while safeguarding personal privacy. However, the process often leads to a decline in the quality of sample data due to a substantial amount of missing encrypted aligned data, and there is a lack of research on how to improve the model learning effect by increasing the number of samples of encrypted aligned data in federated learning. Therefore, this paper integrates the functional characteristics of deep learning models and proposes a Variational AutoEncoder Gaussian Mixture Model Clustering Vertical Federated Learning Model (VAEGMMC-VFL), which leverages the feature extraction capability of the autoencoder and the clustering and pattern discovery capabilities of Gaussian mixture clustering on diverse datasets to further explore a large number of potentially usable samples. Firstly, the Variational AutoEncoder is used to achieve dimensionality reduction and sample feature reconstruction of high-dimensional data samples. Subsequently, Gaussian mixture clustering is further employed to partition the dataset into multiple potential Gaussian-distributed clusters and filter the sample data using thresholding. Additionally, the paper introduces a labeled sample attribute value finding algorithm to fill in attribute values for encrypted unaligned samples that meet the requirements, allowing for the full recovery of encrypted unaligned data. In the experimental section, the paper selects four sets of datasets from different industries and compares the proposed method with three federated learning clustering methods in terms of clustering loss, reconstruction loss, and other metrics. Tests on precision, accuracy, recall, ROC curve, and F1-score indicate that the proposed method outperforms similar approaches. [ABSTRACT FROM AUTHOR]

The detection of underwater fish targets is critical for ecological monitoring and marine biodiversity research. However, underwater fish detection is typically constrained by problems including low image quality and variable underwater surroundings. On behalf of further improving the underwater fish detection accuracy in complex underwater environments, this paper proposes a dual-path (DP) Pyramid Vision Transformer (PVT) feature extraction network named DP-FishNet. The backbone network DP-PVT composed from the PVT Network is made up of two feature extraction paths. The first represents the Vision Transformer path, which extracts global features to enhance the distinction between the foreground and background of underwater images. The second is the convolutional neural network path, which enhances the accuracy of detecting small targets by extracting local features. Additionally, to more effectively utilize the feature information extracted by the network, this paper provides a promising solution to employ the content-aware reassembly of features (Carafe) in the feature pyramid network (FPN). The seesaw loss is utilized as a classification loss to address the problem of unbalanced samples caused by the gap in the number of fish populations. According to the experimental findings, the AP and AP50 of the DP-FishNet are 76.0% and 95.2%, respectively. In comparison to currently available advanced two-stage detection algorithms, the quantity of computation and parameters is reduced by approximately 40%. DP-FishNet strengthens the ability to extract global and local features from underwater images and enhances feature reuse. DP-FishNet can be utilized to detect fish targets in actual and complicated underwater habitats. [ABSTRACT FROM AUTHOR]

The detection of fastener defects is crucial for ensuring the safety and reliability of high-speed train operations. This paper proposes an intelligent algorithm named YOLO-Fastener for detecting fastener defects in ballastless track systems. The proposed YOLO-Fastener incorporates efficient channel and spatial attention mechanisms, enhancing the extraction of crucial features related to fastener defects. Decision regions of the model in identifying fastener defects are visualized through heatmaps. The model is trained and tested on a limited dataset of high-resolution fastener images collected by a ballastless track detection vehicle equipped with 3-D laser devices. The results show that the precision and recall of the proposed model on the test set are 98.33% and 99.15%, which are 1.63% and 4.81% higher than those of the advanced Faster R-CNN model. In terms of fastener detection efficiency, the proposed model is the fastest with an inference time of 10.4 ms, which is an 18.75% improvement over the result of the advanced YOLOv7 model. • The paper proposes an intelligent algorithm named YOLO-Fastener for detecting fastener defects. • A high-resolution dataset of fastener defects is collected by a ballastless track detection vehicle. • The ECSA module has been developed to extract more crucial fastener defect features. • The GhostConv and RepConv modules are employed to reduce model parameters and increase inference speed. [ABSTRACT FROM AUTHOR]

In computed tomography (CT) imaging, artifacts caused by metals can seriously interfere with the accurate judgment, evaluation, and analysis of object structures. Deep learning presented new opportunities for minimizing metal artifacts. The currently proposed supervised methods mainly resort to simulated data for network training, which is difficult to be applied in real scenarios. Unsupervised methods do not completely suppress artifacts and are prone to instability in the corrected results. Therefore, this paper proposes an unsupervised CycleGAN model based on the efficient Transformer for metal artifact reduction in CT images. The encoder and decoder based on the efficient Transformer achieve the extraction and processing of global image content features and artifact features. By correlating global features through attention mechanisms, vital features are generated to enhance the consistency of recovery information. The efficient Transformer framework reduces the computational complexity of the attention mechanism, allowing efficient processing of standard medical CT images with 512 × 512 pixels. Projection preserving loss uses the metal trace outer projection as a fidelity constraint to overcome the problem that unsupervised networks are difficult to train and prone to spurious information. Experimental results indicate that the proposed method outperforms the original CycleGAN and ADN networks in suppressing metal artifacts while avoiding introducing spurious structures. Specifically, the proposed method improved the PSNR of the correction results by 9.23% and reduced the NMAD by 13.21% compared to the original CycleGAN, and improved the PSNR by 5.17% and reduced the NMAD by 5.57% compared to the ADN. [Display omitted] This paper uses the CycleGAN architecture as the base framework to achieve unsupervised reduction of metal artifacts by introducing an efficient Transformer-based generator and a relative discriminator. The generation model consists of a feature extraction and encoding module and a feature decoding module, which consists of two parts: a forward loop and a backward loop. A relative discriminator is used to effectively discriminate and classify the generated image from the real image to more effectively facilitate artifact correction by the generator. The efficient Transformer-based encode enables the extraction and processing of global image content and artifact features and the correlation of global feature relevance through an attention mechanism. This approach produces stronger features to enhance the consistency of the recovered information. Using metal trace outer projection as a projection preserving loss enables more accurate recovery of image information and prevents the generation of spurious structures, further enhancing the fidelity of the recovered information. • Efficient Transformer Unsupervised MAR network effectively extracts CT features. • MAR based on global features improves the consistency of the recovered information. • Efficient Transformer module optimizes the computational efficiency and memory usage. • Projection preserving loss improves the recovery fidelity of image content. [ABSTRACT FROM AUTHOR]

The most frequent chronic illness affecting the elderly and one with a high incidence rate is Alzheimer's disease (AD). Deep Learning (DL) and Machine Learning (ML) techniques has significant success and gained popularity in medical imaging. It has emerged as the method of choice for examining medical images and has drawn considerable interest in identification of AD. The review paper mainly focus on image pre-processing which contributes on noise removal, illumination and intensity correction in Magnetic Resonance (MR) images, segmentation methods which helps in extracting the region of interest for AD detection, feature extraction which are considered as the inputs for classification and various machine and deep learning algorithms are analysed for detecting AD. The survey focus on papers related with academic year 2013 to 2023. The maximum number of paper considered for the review falls from 2019 with 13 papers, 2021 with 27, 2022 with 18 and 2023 with 22 papers and other papers are related with dataset reference and papers published before 2019. From the survey, deep learning techniques are more robust in detecting AD. [ABSTRACT FROM AUTHOR]

The aging of transformer oil–paper insulation (OPI) contributes significantly to transformer failure and reduced service life. Therefore, reliable and precise evaluation of the aging status of transformer insulation condition is crucial in averting transformer failure. This article proposes a novel deep learning-based aging condition assessment technique of transformers by combining FDS (Frequency domain spectroscopy) and squeeze-and-excitation-enabled convolutional neural network (SE-CNN). An intrinsic product-based transformation using a Gramian angular field is employed to dimensionally expand the 1-D FDS signals acquired from OPI samples into 2D images. Each feature extraction channel in the SE-CNN-based aging diagnosis framework is assigned a specific weight, enabling the model to concentrate on the most relevant features while discarding the rest. Furthermore, the effect of moisture variation on the model is also considered under three different conditions to ensure the consequence of compensation for the effect of moisture on FDS. A comparison with traditional machine learning models has also been made to ensure the viability of the proposed methodology. Additionally, the model is interpreted by creating explanation weight plots of the feature coefficients using a local interpretable model-agnostic explanations (LIME) method in Python. The findings demonstrate that the novel aging diagnosis framework visually assesses the transformer's OPI system with the highest accuracy of 99.12% in the test set. Furthermore, it facilitates the automatic extraction of deep aging features, thus providing better generalization and convergence capability. [ABSTRACT FROM AUTHOR]

Automatic detection of Diabetic Retinopathy (DR) is critically important, as it is the primary reason of irreversible loss of vision in the economically active populations in the developed countries. Early detection of the onset of Diabetic Retinopathy can greatly benefit clinical treatment; although several different feature extraction methods have been proposed, the task of retinal image classification remains tedious even for trained clinicians. This paper emphasizes on Diabetic Retinopathy detection as well as the analysis of the different stages of DR, performed on fundus images using Deep Learning algorithms. Fundus images of the patient were provided as input to the developed model evaluated using the real-time dataset of the hospital. The proposed ResNet-18 architecture with swish function has achieved an accuracy of 93.51%, sensitivity of 93.42%, precision of 93.77% and F1-score of 93.59%. The paper concludes with a comparative study of Simple CNN, VGGNet-16, MobileNet-V2 and ResNet architectures and other state-of-art approaches, which highlights ResNet-18 with Swish as the most effective deep learning classifier model for DR detection. • Deep learning outperforms traditional screening for Diabetic Retinopathy, reducing ophthalmologists' workload. • ResNet18's residual connections mitigate the vanishing gradient problem and enable effective retinal image pattern learning. • Swish activation in ResNet18 enhances Diabetic Retinopathy detection with improved accuracy and faster training convergence. [ABSTRACT FROM AUTHOR]

[Display omitted] • A corn quality detection method based on spectral technology and deep learning was proposed. • This method solves the time-consuming and labor-intensive problem of corn quality detection. • This paper provides a new idea and method for corn quality detection. • The research results provide technical support for the rapid detection of crop quality. Corn is an important food crop in the world. With economic development and population growth, the nutritional quality of corn is of great significance to high-quality breeding, scientific cultivation and fine management. Aiming at the problems of cumbersome steps, time-consuming and laborious, and low accuracy in the current research on corn quality detection. This paper proposes to combine near-infrared (NIR) spectroscopy technology with deep learning technology to build a corn quality detection model based on convolutional neural network (LeNet-5). The original spectral data were preprocessed by wavelet transform (WT) and multivariate scattering correction (MSC) to remove noise interference and spectral scattering information. The Competitive Adaptive Reweighted Sampling Algorithm (CARS) was applied to optimize the characteristic wavenumber and reduce redundant data. According to the optimized characteristic wave number, it was input into the constructed corn quality detection model for simulation test, and the average detection accuracy rate of the test set was 96.46%, the average precision rate was 95.42%, the average recall rate was 97.92%, the average F1score was 96.64%, and the average recognition time was 51.95 s. Compared with traditional machine learning models such as BP neural network, K Nearest Neighbor (KNN), Support Vector Machine (SVM), Generalized Linear Model (GLM), Linear Discriminant Analysis (LDA), and Naive Bayesian (NB), the deep learning LeNet-5 network model constructed in this paper has an average accuracy increase of 39.32%, and has a higher detection accuracy. [ABSTRACT FROM AUTHOR]

• In the field of brain-computer interface, automatic recognition of emotions based on electroencephalogram (EEG) signals is of great significance. • At present, deep learning can deeply mine the information in the data, especially the convolutional neural network (CNN) and the recurrent neural network (LSTM), which has remarkably improved the accuracy in numerous fields, hence it is applied by researchers to EEG-based emotional identification research. • Nonetheless, existing CNN and LSTM-based models still rely on data preprocessing and feature extraction. • Furthermore, CNNs have limitations in perceiving global dependencies while LSTMs have problems such as vanishing gradients in the case of long sequences. • In this paper, we proposed the Spatiotemporal Symmetric Transformer Model (STS-Transformer), an effective EEG emotion recognition model, to overcome the current problems. • STS-Transformer is an end-to-end framework that directly recognizes emotion from raw EEG signals without data preprocessing and feature extraction. • This method achieves 89.86% and 86.83% accuracy respectively in the binary classification of valence and arousal in the DEAP dataset; in the DREAMER dataset, the binary classification accuracy of valence and arousal is 85.09% and 82.32%. • Therefore, our method exhibits remarkable advantages over other end-to-end models in similar studies. In the field of brain-computer interface, automatic recognition of emotions based on electroencephalogram (EEG) signals is of great significance. At present, deep learning can deeply mine the information in the data, especially the convolutional neural network (CNN) and the recurrent neural network (LSTM), which has remarkably improved the accuracy in numerous fields, hence it is applied by researchers to EEG-based emotional identification research. Nonetheless, existing CNN and LSTM-based models still rely on data preprocessing and feature extraction. Furthermore, CNNs have limitations in perceiving global dependencies while LSTMs have problems such as vanishing gradients in the case of long sequences. In this paper, we proposed the Spatiotemporal Symmetric Transformer Model (STS-Transformer), an effective EEG emotion recognition model, to overcome the current problems. STS-Transformer is an end-to-end framework that directly recognizes emotion from raw EEG signals without data preprocessing and feature extraction. This method achieves 89.86% and 86.83% accuracy respectively in the binary classification of valence and arousal in the DEAP dataset; in the DREAMER dataset, the binary classification accuracy of valence and arousal is 85.09% and 82.32%. Therefore, our method exhibits remarkable advantages over other end-to-end models in similar studies. [ABSTRACT FROM AUTHOR]

Due to the high altitude, high imaging resolution, complex background, and tilted shooting angle of the UAV images, detecting small targets poses challenges. In recent years, numerous deep learning-based methods have achieved good results in feature extraction and localization classification, but they fail to meet the needs of faster and more accurate recognition of small objects. This paper proposes a novel UAV remote sensing image small target detection algorithm called STF-YOLO to address the aforementioned issues. Firstly, we introduce the innovative SwinTransformer, a popular technique in the field of NLP, and combine it with CNNs to propose a novel convolutional structure called STRCN. Furthermore, this paper presents a novel lightweight classifier called CNeB, which is innovatively designed and incorporated into the Backbone part. CNeB exhibits advantages such as high recognition accuracy and parameter efficiency. During the process, we further enhance the prediction accuracy by incorporating a parameter-free attention mechanism called SimAM to improve the overall precision of the model. Moreover, we propose a novel detection head called CWDHead, which strengthens the weighting of classification capabilities and significantly improves the recognition accuracy of similar inter-class and intra-class instances. Lastly, we design a novel data augmentation method called SVM, which enriches the training dataset and enhances the robustness and generalization ability of the model. The results indicate that by applying enhancements such as WBF and TTA, our proposed algorithm, STF-YOLO, exhibits significant improvements over current state-of-the-art methods on the publicly available VisDrone dataset. It demonstrates noticeable advancements in widely recognized metrics such as 3.9%mAP, and 2.0%AP50. [ABSTRACT FROM AUTHOR]

• Ktrans as a priori information is beneficial for the segmentation of NPC in DCE-MRI. • A network model MBRU-Net for multi-branch feature extraction is proposed. • Multi-branch networks outperform single-branch networks. Accurate segmentation of nasopharyngeal carcinoma (NPC) lesion areas from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) facilitates subsequent diagnostic and prognostic workups. However, in previous studies, little consideration has been given to incorporating the pharmacokinetic feature Ktrans as auxiliary information for segmenting NPC in DCE-MRI. Therefore, this paper proposes the use of a pharmacokinetic extended Tofts and Kermode (ETK) model to obtain the Ktrans feature from DCE-MRI and combine it with MRI images for nasopharyngeal tumor segmentation. Additionally, this paper proposes a multi-input branch residual U-Net (MBRU-Net) model that effectively fuses DCE-MRI features and Ktrans features. The effectiveness of the multibranch network is validated by comparing MBRU-Net with ResU-Net with DCE-MRI + Ktrans data. Additionally, different models are trained with DCE-MRI and DCE-MRI + Ktrans data separately and compared to validate the effectiveness of multimodal data using the Dice coefficient. Our proposed MBRU-Net achieves the best Dice in this study (67.39 ± 15.79%), higher than ResU-Net's Dice (65.57 ± 17.52) based on DCE-MRI and Ktrans data. U-Net, SegNet, R2U-Net, and ResU-Net achieve better results in terms of segmenting tumor regions with DCE-MRI + Ktrans data than those of the corresponding models with DCE-MRI data alone, where U-Net has the best performance (DCE-MRI + Ktrans: DCE-MRI = 66.31 ± 17.80%: 61.10 ± 24.14%). The results show that it is beneficial to add a pharmacokinetic parametric (Ktrans) map as prior information to the conventional anatomical MRI-based segmentation task, and multibranch network structures perform better than single-branch network structures in terms of NPC segmentation. [ABSTRACT FROM AUTHOR]

Industry 4.0 is of great significance for the development of oil industry. One of the pivotal steps towards achieving oil Industry 4.0 is accurately mastering the oilfield production dynamics, especially the changes of process parameters. Among many process parameters, the accurate modeling of water cut is extremely critical and difficult. In this paper, under laboratory conditions, oil–water flows with various preset water cuts are simulated. A fluid sensor, equipped with eight concave-shaped conductive electrodes, is employed to capture multi-channel measurement data, continuously recording the oil–water flow process from multiple angles. Subsequently, a novel flow behavior-heuristic deep learning model, named FBHWC model, is developed to model the relationship between the measurement data and water cut, achieving water cut measurement. The FBHWC model is guided by the complex flow behaviors of oil–water flows and consists of two key modules. Particularly, the multi-level feature fusion module focuses on the high-order feature extraction and fusion of sensor measurement data, while the multi-scale measurement module uses fully convolutional design to achieve accurate measurement of water cut. Experimental results show that the FBHWC model has excellent performance in measuring water cut, with mean square error of 0.016%. All these open up a new venue for exploring industrial multiphase flows through combining multi-electrode sensor and deep learning. [ABSTRACT FROM AUTHOR]