Search

Intrusion Detection System (IDS) can ensure the network security by identifying network intrusions according to the abnormal traffic data. However, the intrusion detection data has the problem of high dimensionality and changes with network and attack environments, which leads to the poor performance and poor portability of intrusion detection algorithms. Therefore, this paper proposes an intrusion detection algorithm based on joint symmetric uncertainty and hyperparameter optimized fusion neural network. Firstly, a feature selection method based on symmetric uncertainty and approximate Markov blanket is proposed, which fully considers the correlation and redundancy of features, and also the correlation between combined features and the class label, so as to reduce the data dimensionality. Secondly, the CNN-LSTM classifier fused with Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) is used to extract the spatial features and temporal features to improve the classification performance. Finally, the Particle Swarm Optimization (PSO) algorithm is improved and used to automatically optimize the hyperparameters of the classifier, so that the classifier can be applied to different intrusion detection datasets with better generalization ability and portability. Experiments have verified the effectiveness and superiority of the proposed algorithm on multiple evaluation indicators. • An effective algorithm for intrusion detection is proposed in this paper. • Feature selection is based on symmetric uncertainty and approximate Markov blanket. • A fusion neural network is constructed to extract the spatial and temporal features. • The PSO algorithm is improved to automatically optimize the hyperparameters. [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]

In this paper, we propose a new deformable convolutional network with attention mechanism to deal with the Chinese character of various dynasties. These ancient Chinese characters are hieroglyph with special spatial structure and gradually evolving into modern Chinese characters. However, because of the intrinsic limitations of traditional convolutional networks in the model, the geometric structure of the convolutional module is fixed. Therefore, we build a new architecture called attention-based deformable networks for this task. Besides, we construct a new dataset called the Chinese characters from Various Dynasties Dataset (CCDD), which includes the evolution of Chinese characters between major dynasties. Deformable convolution based on attention mechanism can obtain those more important offsets, resulting in better performance than the original deformable convolution. In the experimental stage, extensive experiments validate the performance of our approach. Firstly, the effectiveness of this network was verified on two text datasets, Mnist(99.38%) and notMnist(98.22%) both comparing with several modules. Finally, the best performance was achieved on the CCDD dataset(91.50%) compared to other classical deep learning networks. This is the first paper to apply deep learning methods to recognize and classify the characters of major dynasties in China. And archaeological workers need to be able to recognize Chinese characters from various periods in China, and our algorithm can assist them in completing this task to a certain extent. It can also help cultural relic enthusiasts roughly determine the age of cultural relics. [ABSTRACT FROM AUTHOR]

Marine ecosystems and their fish habitats are becoming increasingly important due to their integral role in providing a valuable food source and conservation outcomes. Due to their remote and difficult to access nature, marine environments and fish habitats are often monitored using underwater cameras to record videos and images for understanding fish life and ecology, as well as for preserve the environment. There are currently many permanent underwater camera systems deployed at different places around the globe. In addition, there exists numerous studies that use temporary cameras to survey fish habitats. These cameras generate a massive volume of digital data, which cannot be efficiently analysed by current manual processing methods, which involve a human observer. Deep Learning (DL) is a cutting-edge Artificial Intelligence (AI) technology that has demonstrated unprecedented performance in analysing visual data. Despite its application to a myriad of domains, its use in underwater fish habitat monitoring remains under explored. In this paper, we provide a tutorial that covers the key concepts of DL, which help the reader grasp a high-level understanding of how DL works. The tutorial also explains a step-by-step procedure on how DL algorithms should be developed for challenging applications such as underwater fish monitoring. In addition, we provide a comprehensive survey of key deep learning techniques for fish habitat monitoring including classification, counting, localisation, and segmentation. Furthermore, we survey publicly available underwater fish datasets, and compare various DL techniques in the underwater fish monitoring domains. We also discuss some challenges and opportunities in the emerging field of deep learning for fish habitat processing. This paper is written to serve as a tutorial for marine scientists who would like to grasp a high-level understanding of DL, develop it for their applications by following our step-by-step tutorial, and see how it is evolving to facilitate their research efforts. At the same time, it is suitable for computer scientists who would like to survey state-of-the-art DL-based methodologies for fish habitat monitoring. • Marine environments and fish habitats are widely surveyed using underwater cameras. • Deep learning (DL) offers an effective solution for processing underwater videos. • We write a tutorial on the key concepts and steps of DL for underwater monitoring. • We provide a survey of key DL-based visual fish monitoring methods/datasets. • We shed light on challenges/opportunities of DL for underwater fish monitoring. [ABSTRACT FROM AUTHOR]

Edge detection is of critical importance for middle-level and high-level tasks in computer vision. Existing edge detection methods usually use VGG16 as the encoding network and achieve exceptional performance through transfer learning, which has the characteristics of high parameters and high computational cost. Researchers have implemented edge detection by designing decoding networks. Unlike existing methods, this paper is inspired by the effective mechanisms of edge detection in the parallel pathway of biological vision and proposes a new lightweight encoding–decoding structure for edge detection networks, which we refer to as XYW-Net. In the encoding network, we drew inspiration from the receptive field properties of X-type cells, Y-type cells, and W-type cells involved in the parallel pathway, and developed more meticulous convolutional models. Based on the structure of the parallel pathway, the three cell modules are combined into a novel encoding network and excellent feature extraction performance is obtained. Within the decoding network, this paper draws inspiration from the feature integration ability of the inferotemporal cortex (IT) and introduces a novel feature integration module to constitute the decoding network. Experiments show that the proposed network in this paper obtains the Optimal Dataset Scale (ODS) = 0.812 on the BSDS500 dataset with only 0.79 M parameters. Additionally, the model has exhibited outstanding ODS performance on various publicly available edge detection datasets, including NYUD, BIPEDv1, and Multicue. Moreover, future research could concentrate on implementing more efficient antagonistic mechanisms and devising networks grounded in higher-level visual cortical mechanisms. This could further augment the performance and efficiency of edge detection networks. We encourage researchers to delve into these directions within the field. The codes are available at https://github.com/PXinTao/XYW-Net. [ABSTRACT FROM AUTHOR]

• Pinch force strength is accurately predicted using sEMG. • Different pinch movements are accurately classified using sEMG. • Similarity of EMG features is evaluated. • An EMG feature-to-image method is proposed. The movement of intelligent EMG-driven prosthesis mainly relies on the synergy of different fingers to achieve function of grasping objects. The paper proposes a novel scheme for force prediction and movement classification about pinch between different fingers based on surface electromyography (sEMG) using machine learning. The pinch force and sEMG signals are recorded synchronously by a data acquisition device. Eight features are extracted, which are proven to have better performance in the estimation of sEMG-to-force. We present a novel feature selection method that uses the one-dimensional time series similarity assessment based on Manhattan distance to eliminate the repetitive information between features. Three optimal features carrying less repetitive information are retained. Seven machine learning algorithms are used to predict force strength. The results show that the Long Short Term Memory (LSTM) has the best performance of force prediction, achieving a R 2 of 0.9517 and RMSE of 3.2723. The paper proposes a novel method of converting EMG feature sequences to the normalized gray image in order to classify the finger movement. Five classifiers based on image feature extraction and the Convolutional Neural Network (CNN) are developed respectively. The experimental results indicate that the CNN performs best, achieving an accuracy of 97.66%. In this way, it not only realizes the accurate force prediction, but also realizes the movement classification between different fingers. The proposed methodology has the potential to realize simultaneous force and movement control of prosthetic hand. [ABSTRACT FROM AUTHOR]

Fiber-optic hydrophones based on optical fiber sensing and communication technology have developed rapidly in recent years. With the advantages of high sensitivity, small size, and large dynamic range, they are easy to detect underwater acoustic signals, showing great potential in the field of marine ecological protection. Researchers can monitor the marine ecological environment by recognizing underwater optical fiber sensing signals collected by fiber-optic hydrophones in real-time. Deep learning models can achieve high recognition accuracy, but they usually have large amounts of parameters that makes them difficult to be deployed in mobile terminals of marine survey ships. To solve the key problem, this paper discussed the design of the Deep Light-Weight Attention Residual Network (DLA-ResNet). It introduces the depth-wise separable convolution (DSC) and the convolutional block attention module (CBAM) into ResNet18 to make its performance better. ResNet18 is a two-dimensional convolutional network (CNN) model applying residual blocks. Experiments verify that the accuracy of DLA-ResNet improved from ResNet18 is increased by about 2.04% and its size is reduced by 85.67%. The proposed scheme in this paper realizes an average recognition rate of 96.50% for eight types of underwater optical fiber sensing signals in the application of marine ecological environment monitoring. Moreover, its model size is only 6.12 MB and it only takes 27.90 ms, which is available to meet the needs of reliability and effectiveness. [ABSTRACT FROM AUTHOR]

The 2019 coronavirus disease outbreak, caused by the severe acute respiratory syndrome type-2 virus (SARS-CoV-2), was declared a pandemic in March 2020. Since its emergence to the present day, this disease has brought multiple countries to the brink of health care collapse during several waves of the disease. One of the most common tests performed on patients is chest x-ray imaging. These images show the severity of the patient's illness and whether it is indeed covid or another type of pneumonia. Automated assessment of this type of imaging could alleviate the time required for physicians to treat and diagnose each patient. To this end, in this paper we propose the use of Convolutional Neural Networks (CNNs) to carry out this process. The aim of this paper is twofold. Firstly, we present a pipeline adapted to this problem, covering all steps from the preprocessing of the datasets to the generation of classification models based on CNNs. Secondly, we have focused our study on the modification of the information fusion processes of this type of architectures, in the pooling layers. We propose a number of aggregation theory functions that are suitable to replace classical processes and have shown their benefits in past applications, and study their performance in the context of the x-ray classification problem. We find that replacing the feature reduction processes of CNNs leads to drastically different behaviours of the final model, which can be beneficial when prioritizing certain metrics such as precision or recall. • Convolutional Neural Networks are effective for detecting SARS-CoV-2 on x-ray images. • Intermediate extracted features can be fused through different aggregation functions. • Grouping functions are harder to train but can offer better classification accuracy. • An early restart policy eases training models which make use of grouping functions. [ABSTRACT FROM AUTHOR]

The vulnerability of RGB-based human action recognition in complex environment and variational scenes can be compensated by skeleton modality. Therefore, action recognition methods fusing RGB and skeleton modalities have received increasing attention. However, the recognition performance of the existing methods is still not satisfactory due to the insufficiently optimized sampling, modeling and fusion strategy, even the computational cost is heavy. In this paper, we propose a Dense-Sparse Complementary Network (DSCNet), which aims to leverage the complementary information of the RGB and skeleton modalities at light computational cost to obtain the competitive action recognition performance. Specifically, we first adopt dense and sparse sampling strategies according to the advantages of RGB and skeleton modalities, respectively. And then, we use the skeleton as guiding information to crop the key active region of the persons in the RGB frame, which largely eliminates the interference of the background. Moreover, a Short-Term Motion Extraction Module (STMEM) is proposed to compress the densely sampled RGB frames to fewer frames before feeding them into the backbone network, which avoids a surge in computational cost. And a Sparse Multi-Scale Spatial–Temporal convolutional neural Network (Sparse-MSSTNet) is designed to modeling sparse skeleton. Extensive experiments show that our method effectively combines complementary information of RGB and skeleton modalities to improve recognition accuracy. The DSCNet achieves competitive performance on NTU RGB+D 60, NTU RGB+D 120, PKU-MMD, UAV-human, IKEA ASM and Northwest-UCLA datasets with much less computational cost than exiting methods. The code is available at https://github.com/Maxchengqin/DSCNet. • Targeted sampling strategies for RGB and skeleton. • A network for sparse skeleton-based action recognition is proposed. • Segmentally aggregate the motion information in dense RGB frames. • Achieves state-of-the-art performance with much less computational cost. [ABSTRACT FROM AUTHOR]

Seamless positioning ability has become an essential requirement in large-scale smart city scenes with the development of Artificial Intelligence of Things technology. The performance of seamless positioning is limited by the inaccurate crowdsourced navigation database, cumulative error of built-in sensors, and changeable measurement errors of different location sources. In order to solve these problems, this paper presents the CrowdLOC-S framework, which provides a concrete and accurate indoor/outdoor localization performance using the combination of crowdsourced Wi-Fi fingerprinting, Global Navigation Satellite System (GNSS), and low-cost sensors. A data and model dual-driven based trajectory estimator is developed for improving the long-term positioning performance of built-in sensors, and a hybrid one-dimensional convolutional neural network (1D-CNN), Bi-directional Long Short-Term Memory (Bi-LSTM), and Multilayer Perceptron (MLP) enhanced quality indicator is proposed for quality evaluation of crowdsourced trajectories and further Wi-Fi fingerprinting database construction. Besides, the transfer learning approach is applied in the quality indicator for autonomously predicting the location errors towards different indoor and outdoor location sources and realizing seamless scenes switching. Finally, a unified extended Kalman filter is developed to realize multi-source integration-based seamless localization using the positioning information provided by indoor and outdoor location sources and corresponding quality indicator results. Comprehensive experiments demonstrate that the presented CrowdLOC-S system is proven to realize precise and efficient indoor and outdoor positioning performance in complex and large-scale urban environments. [ABSTRACT FROM AUTHOR]

Detecting bearing faults helps ensure the healthy operation of machinery and prevents serious accidents. However, fault diagnosis method based on deep learning relies on the correlation between the extracted vibration signal features. It does not consider the causal relationship between fault, noise, and vibration waveform changes, resulting in lower bearing fault diagnosis accuracy under realistic working conditions. This study proposes a graph neural network (GNN) method based on the Granger causality test for bearing fault detection called GCT-GNN to address this issue. The proposed method first performs feature transformation on the original signals to extract time-domain and frequency-domain features, forming a feature matrix. Then, spectral analysis is conducted on both faulty signals with noise and healthy signals to calculate the lag order between them. Subsequently, the Granger causality test is employed to quantify the impact of faults and noise on signal changes, and the quantified results are used to calculate weights, constructing an adjacency matrix. The completed adjacency matrix and feature matrix are input into the GNN for feature mapping, last classifying the bearing fault data. In this paper, five models, Deep residual shrinkage Network (DRSN), GNN, Support Vector Machines(SVM), Random Forests (RF), and Convolutional Neural Network(CNN) were selected, and comparative experiments were carried out on two public data sets. The results show that, compared with other models, GCT-GNN has better anti-noise ability and is better than other models in various fault diagnoses. • The effect of noise on vibration signal is quantified by Granger causality test. • The lag order was calculated by using spectrum analysis method. • The method of calculating weight by causality is proposed to construct causality graph. • Construct a graph neural network using causal graphs. [ABSTRACT FROM AUTHOR]

This paper presents an innovative deployment strategy for a power-aware and delay-aware TinyML model optimized for agricultural soil quality monitoring and management. The proposed method employs an enhanced fusion of Dynamic Voltage and Frequency Scaling (DVFS), Sleep/Wake Strategies based on Genetic Algorithms, Energy Harvesting, and Task Partitioning. Through this method, the model reduces energy consumption by 8.5% and delay by 10.4%, while maintaining accuracy comparable to state-of-the-art techniques such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Q Learning techniques. The proposed model also uses a Proof-of-Energy Efficiency (PoEE) based consensus for creation of new blocks. These blocks are added to different sidechains for strengthening the security levels. While blockchains require higher energy during mining, the proposed PoEE Model uses a light weight consensus, which is backed by Q Learning based sidechain formation process. This lightweighted combination allows the model to reduce delay, maintain lower energy consumption, with better communication throughput even under multiple attack scenarios. This model is necessary due to the growing demand for sustainable and accurate agricultural practices. The quality of the soil is crucial to soil yield and productivity. Monitoring and managing soil conditions enables farmers to optimize irrigation, fertilization, and overall resource allocation, leading to an increase in soil yield and a decrease in environmental impacts. However, the deployment of energy-efficient and low-latency TinyML models is essential to enable real-time decision-making in agricultural settings with limited resources. This power-aware and delay-aware TinyML model has applications in various agricultural real-time scenarios. It can be used to continuously monitor soil quality parameters including moisture content, pH level, and nutrient levels for various farm types. Incorporating sensors and low-power microcontrollers, the model enables on-site analysis and timely farmer feedbacks. The decreased energy consumption extends the battery life of deployed devices, whereas the decreased delay ensures that soil management interventions and treatments are responsive and timely for different scenarios. [ABSTRACT FROM AUTHOR]

Retrieving similar images is becoming an urgent need for us with the continuous growth of large-scale data. However, whether the dominant image retrieval methods are Convolutional Neural Networks (CNNs) or the recently emerging Visual Transformer (ViT), their complex computation, insufficient feature extraction, and mismatched weights greatly influence the efficiency and retrieval accuracy. In this paper, we propose a Pooling-based Visual Transformer with low complexity attention hashing (PTLCH) for image retrieval. First, a backbone network for Pooling-based Vision Transformer (PiT) feature learning is designed to combine the pooling in CNN and the ViT to achieve the purpose of spatial dimensionality reduction while learning rich semantic information. Second, a low complexity attention (LCA) module is incorporated into PiT, which works by combining the positional deviation with the key matrix and the value matrix and then matrix multiplying with the query matrix. LCA explores rich contextual information to enable network learning of more granular feature information. Finally, a new loss framework is proposed where we focus on the effect of difficult and erroneous samples on accuracy. By using different improved cross-entropy losses, better weights are assigned to the learning samples of our network, which effectively improves learning hash coding. We have conducted extensive experiments on three public datasets, CIFAR-10, ImageNet100, and MS-COCO, which have the highest mean average precision of 93.76%, 92.62%, and 90.60%, respectively. [Display omitted] • An end-to-end learning framework for image retrieval is proposed, namely PTLCH. • LCA modules are embedded on top of the network. • A new deep hash learning loss framework is designed. • The superior performance of PTLCH method is demonstrated experimentally. [ABSTRACT FROM AUTHOR]

For achieving better performance, the majority of deep convolutional neural networks have endeavored to increase the model capacity by adding more convolutional layers or increasing the size of the filters. Consequently, the computational cost increases proportionally with the model capacity. This problem can be alleviated by dynamic convolution. In the case of pixel-level labeling, existing pixel-level dynamic convolution methods have a smaller scanning area than ordinary convolution or image-level dynamic convolution and are thus unable to exploit fine contextual information. As a consequence, pixel-level dynamic convolution is more sensitive to large-scale varying objects and confusion categories. In this paper, we propose a scale-pyramid dynamic atrous convolution (SDAConv) and exploit multi-scale pixel-level features in finer granularity, in order to efficiently increase model capacity, exploring contextual information, capture detail information and alleviate large-scale variation problem at the same time. Through kernel engineering (instead of network engineering), SDAConv dynamically arranges atrous filters in the individual convolutional kernels over different semantic areas at dense scales in the spatial dimension. By simply replacing the regular convolution with SDAConv in SOTA architectures, extensive experiments on three public datasets, Cityscapes, PASCAL VOC 2012 and ADE20K benchmarks demonstrate the superior performance of SDAConv on pixel-level labeling tasks. • A new scale-pyramid dynamic convolution(SDAConv) is proposed. • SDAConv is truly dynamic and multiple receptive fields. • SDAConv can improve localization accuracy and explore rich contextual information. • SDAConv can be used in a plug-and-play fashion in any architecture. [ABSTRACT FROM AUTHOR]

With the rapid advancement of multimedia technologies, there is a growing demand for reliable methods to verify image integrity. Camera model identification, a passive approach aiming to determine the specific capturing device model, has garnered considerable attention in the field of source camera forensics. In this paper, we first propose a novel patch selection method that enhances the diversity of training data by utilizing the uniform local binary pattern operator to reveal spatial textual information. Secondly, we introduce a complex dual-path enhanced ConvNeXt network for camera model identification, effectively leveraging the multi-frequency information present in the image. Notably, our network demonstrates the ability to learn camera model-related features without relying on a residual prediction module. Finally, extensive experimental results on both Dresden and Vision datasets shown that the proposed network outperforms several state-of-the-art methods in both teams of identification accuracy and computational efficiency. • Proposing a patch selection method of the uniform local binary pattern operator. • Introducing a dual-path enhanced ConvNeXt network for camera model identification. • Obtaining robust results on the Dresden and Vision datasets. [ABSTRACT FROM AUTHOR]

Thunderstorms can be a large source of disruption for European air-traffic management causing a chaotic state of operation within the airspace system. In current practice, air-traffic managers are provided with imprecise forecasts which limit their ability to plan strategically. As a result, weather mitigation is performed using tactical measures with a time horizon of three hours. Increasing the lead time of thunderstorm predictions to the day before operations could help air-traffic managers plan around weather and improve the efficiency of air-traffic-management operations. Emerging techniques based on machine learning have provided promising results, partly attributed to reduced human bias and improved capacity in predicting thunderstorms purely from numerical weather prediction data. In this paper, we expand on our previous work on thunderstorm forecasting, by applying convolutional neural networks (CNNs) to exploit the spatial characteristics embedded in the weather data. The learning task of predicting convection is formulated as a binary-classification problem based on satellite data. The performance of multiple CNN-based architectures, including a fully-convolutional neural network (FCN), a CNN-based encoder–decoder, a U-Net, and a pyramid-scene parsing network (PSPNet) are compared against a multi-layer-perceptron (MLP) network. Our work indicates that CNN-based architectures improve the performance of point-prediction models, with a fully-convolutional neural-network architecture having the best performance. Results show that CNN-based architectures can be used to increase the prediction lead time of thunderstorms. Lastly, a case study illustrating the applications of convection-prediction models in an air-traffic-management setting is presented. • CNNs are applied to numerical weather prediction data to predict thunderstorms. • CNNs outperform MLP architectures in the task of predicting thunderstorms. • ML-based weather forecasts allow for better planning of air traffic operations. [ABSTRACT FROM AUTHOR]

Automatic modulation recognition (AMR) has garnered significant attention in both civilian and military domains, with applications ranging from spectrum sensing and cognitive radio (CR) to the deterrence of adversary communication. Index modulation (IM) represents an innovative digital modulation technique that exploits the indices of parameters of communication systems to transmit extra information bits. This paper aims to examine the performance of a convolutional neural network (CNN)-based AMR across various IM systems, including spatial modulation (SM), quadrature spatial modulation (QSM), and generalized spatial modulation (GSM) with eight digital modulation schemes. In this study, we leverage confusion matrices, receiver operating characteristic (ROC) curves, and F1 scores to illustrate the recognition model's outputs. • Modulation recognition models exploit deep learning algorithms effectively. • Index modulation is a novel technique for conveying additional information. • Three datasets are created for the recognition of eight digital modulation schemes. • Three modulation recognition scenarios analyze the system's performance. • Quadrature amplitude modulation types are more robust than phase shift keying types. [ABSTRACT FROM AUTHOR]

Deep learning-based automatic sleep staging methods have been widely applied for sleep scoring and sleep diagnosis. However, most methods consider only a single temporal scale when dealing with sleep signals. Furthermore, these methods are limited to target only a single-age group or single dataset. In this paper, we propose a multi-scale temporally focused sleep staging model, MAGSleepNet, which can be used for multi-age groups simultaneously. MAGSleepNet consists of (1) a group age classification (GAC) module that can offer a preliminary screening on epidemic estimation of multiple age, (2) a dimensional expansion module (DEM) that can expand the dimension of the input signals, (3) a sequential multi-scale convolutional neural network (SMCNN) that extracts multi-scale features and short-time temporal information, and (4) sequence temporal encoder (STE) that extracts sequential temporal information. In addition, two auxiliary tasks are used to complement the short-time temporal information and to reassign the probabilities of different age groups to enhance the model robustness, respectively. The MAGSleepNet is evaluated in adult, child and infant, tested on MASS, CHAT, and CHFU datasets with accruacy of 86.7%, 80.1% and 66.5%, outperforming the state-of-the-art methods. Based on the excellent performance of the proposed method, it is expected to pave the way for automatic sleep staging methods with strong generalizability for multiple age groups. • A novel model is proposed for simultaneous sleep staging in multi-age groups. • The hierarchical architecture of model ensures the effective feature extraction. • An auxiliary task is used to provide temporal information on different scales. • Numerous experiments are conducted to evaluate the proposed model. [ABSTRACT FROM AUTHOR]

The complexity of the environment makes rapidly-exploring random tree (RRT) difficult to handle dynamic obstacle avoidance and system constraint in real-time path planning for autonomous vehicles. To handle this issue, this paper proposes a novel real-time double-tree rapidly-exploring random tree (RDT-RRT) algorithm framework. The collision-free path by RRT after B-spline smooth treatment is adopted as the reference path to reduce invalid sampling. Integrating G1 Hermite interpolation with G2 Hermite interpolation reduces the sampling dimension and takes more efficient samples. The optimal distance metric is designed considering dynamic collision detection mechanism and utilized to estimate the costs of the samples in terms of path curvature. Moreover, to have a better understanding of the environment, convolutional neural network (CNN) is embedded to strengthen the collision detection mechanism. By RDT-RRT the smooth, collision-free paths with small curvature changes can be evaluated. For the evaluations of our proposal in global and local planning, the experiments for a real scaled autonomous vehicle are implemented through parallel computing. By comparing with the mainstream RRT-based algorithms, it has shown that in terms of the path quality, our method reduces 92 % and 88 % of cumulative curvature change respectively in obstacle-free and static obstacle scenarios. Compared with other RRT methods, RDT-RRT performs faster convergence rate and Parallel computing increases the updating frequency from 1.1 Hz to 5.5 Hz. The obstacle avoidance capabilities are also improved. [ABSTRACT FROM AUTHOR]

With the advent of high-resolution surveillance equipment, fire detection aimed at obtaining more detailed information has drawn considerable attention. The methods are based on convolutional neural networks (CNNs), which have been widely applied to automatically extract fire detection image features. However, the existing CNN-based fire detection methods are only designed for fixed-scale images. Thus, these methods are still difficult to use for fire detection due to the scale variation in the fire object and are infeasible for satisfying the requirement of various hardware of different scale images. In this paper, a fire disaster detection method that can deal with varied-scale images is proposed. First, the dense connection is used to enhance the information flow between different layers. Then, the groups channel attention is utilized to recalibrate the features. Finally, multiscale spatial feature pooling is employed to fuse different scale features. Specifically, the module allows us to predict different scale images. Experimental results demonstrate that the proposed method achieves 91.4 accuracy using fixed scale training, and 92.4 accuracy using multiscale training. [ABSTRACT FROM AUTHOR]

In the construction of intelligent transportation systems, short-term traffic flow prediction is of great significance for the advancement of traffic network management. But due to the presence of many complex factors in both spatial and temporal domains, it remains a complex and challenging task. Existing literature usually employs the convolutional neural network (CNN)-based methods in capturing spatiotemporal correlations. These CNN-based methods often use a single-channel data structure to represent different periodic patterns, which makes the model susceptible to over-parameterization when capturing periodic dependencies and prone to information loss after convolution. To overcome these limitations, this paper presents a hybrid deep learning method for short-term traffic flow prediction. In this method, a video-shaped multi-channel data structure is designed to represent different periodic patterns more efficiently. Next, a new mixed-pointwise convolution is introduced for capturing periodic dependencies without the negative impacts mentioned above. Lastly, an improved channel attention mechanism is proposed to learn channel inter-dependencies with controllable parameter usage. The proposed method is lightweight, yet highly effective. Compared to the state-of-the-art baseline method, it reduces the root mean squared error by up to 6.7% on the PeMSD4 dataset and 13.3% on the PeMSD7 dataset, while also achieving substantial improvement in two additional metrics, exhibiting strong robustness and great scalability across various settings. [ABSTRACT FROM AUTHOR]

Developing 3D Convolutional Neural Networks (CNNs) for medical image segmentation is challenging due to the limited number of available labeled medical images and computational resources. Effective design of a 3D CNN requires a better structure and appropriate hyperparameters. Manually designing a neural network and choosing appropriate hyperparameters requires extensive knowledge, a time-consuming and challenging process, resulting in complex and over-parameterized networks. Neuroevolution is an automatic and effective approach to finding an optimal network structure and hyperparameters. This paper proposes an evolutionary based method using Chameleon Swarm Algorithm (CSA) to develop 3D CNN networks for volume segmentation. We also design a search space to build optimal 3D block-based encoder–decoder structure networks for medical image segmentation. We evaluated the proposed method on two different datasets (CT Spleen and MRI Heart) and compared the results of the obtained model with state-of-the-art models. Furthermore, the comparative results show that our proposed method yields 3D CNN networks with superior performance and fewer parameters than other state-of-the-art models utilized in 3D medical image segmentation. [Display omitted] • Proposed a dynamic 3D CNN for medical image segmentation with fewer parameters. • Implemented search space with various blocks, parameters for encoder–decoder network. • Employed CSA to search the optimal network structure and training hyperparameters. • Results demonstrate the effectiveness of proposed method for 3D segmentation tasks. [ABSTRACT FROM AUTHOR]

Understanding the attitudes of users about different topics through their huge amount of comments and opinions on social networks is an emerging hot issue. Analyzing the sentiments of these posts and comments provides useful information utilized for many applications. Multimodal sentiments often contain more information than single-modal sentiments, including text, image, audio and video, which leads to better performance of multimodal sentiment analysis (MSA) compared to single-modal sentiment analysis (SSA). In this paper, a capsule network-based deep ensemble transfer learning approach, called DSY-ETL-MSA, is proposed for MSA on images and texts, the results of which are fused using Yager theory. Capsule networks and ensemble learning methods improve classification performance, and the deep transfer learning approach reduces the training time. A hybrid deep architecture is used for automatic feature extraction in the proposed method. For analyzing the sentiment of image modality, a pre-trained VGG16 model, fine-tuned on the datasets, is used to extract high-level features for image classification. A capsule convolutional neural network (CNN) is also separately used for extracting image features and classifying them. In the text modality, the pre-trained GloVe model is exploited to embed words and 2 separate classifiers are employed for text classifications. The Yager fusion rules are finally used for early and late fusions of the results of the classifiers. The results of the text and image classifiers are combined separately as the early fusion and the final results of them are fused at the decision level as the late fusion. The proposed model is empirically evaluated on the MVSA and T4SA datasets. The significant performance improvement compared to a variety of former methods is shown in the experimental results. The final accuracies obtained by the proposed method on the MVSA and T4SA datasets are 0.9866 and 0.9996, respectively. [ABSTRACT FROM AUTHOR]

Development of human machine interface has become a necessity for modern day machines to catalyze more autonomy and more efficiency. Gaze driven human intervention is an effective and convenient option for creating an interface to alleviate human errors. Facial landmark detection is very crucial for designing a robust gaze detection system. Regression based methods capacitate good spatial localization of the landmarks corresponding to different parts of the faces. But there are still scope of improvements which have been addressed by incorporating attention. In this paper, we have proposed a deep coarse-to-fine architecture called LocalEyenet for localization of only the eye regions that can be trained end-to-end. The model architecture, build on stacked hourglass backbone, learns the self-attention in feature maps which aids in preserving global as well as local spatial dependencies in face image. We have incorporated deep layer aggregation in each hourglass to minimize the loss of attention over the depth of architecture. Our model shows good generalization ability in cross-dataset evaluation and in real-time localization of eyes. • Deep architecture build on stacked hourglass backbone for localization of eye regions. • Self-attention module between hourglasses to learn local spatial and global attention. • Designed a deep layer aggregation module to learn feature dependencies across network. • Differentiable soft-argmax to make the framework an end-to-end trainable architecture. • Precise and fast localization of eyes tested in real-time with varied illumination. [ABSTRACT FROM AUTHOR]

During recent years, deep convolutional neural networks (CNNs) have become increasingly popular in sensor-based human activity recognition (HAR), due to powerful feature extraction capacity. However, the required high computational cost has hindered their practical deployment in real-world applications. To address this issue, this paper presents a novel approach named ABCSearch to prune channels through an artificial bee colony (ABC) algorithm, which is a meta-heuristic optimization algorithm inspired by the foraging behavior of honeybees. Unlike traditional pruning strategies that prioritize channel importance, the proposed method seeks to identify the optimal number of channels at each layer, i.e., the optimal pruning structure. To lessen human interference, we transform the search problem of the optimal pruning structure into a simple optimization problem and solve it automatically through the ABC algorithm, which can significantly decrease both model size and computational cost without compromising performance. Specifically, we validate the effectiveness of ABCSearch for a standard five-layer CNN backbone architecture, which achieves 97.05%, 98.45%, 91.71%, 78.89% classification accuracies on four HAR benchmarks, i.e., UCI-HAR, WISDM, PAMAP2, UniMiB-SHAR. Our approach can yield 87.44%, 89.41%, 89.06%, 89.59% FLOPs reductions, and 89.54%, 88.70%, 88.38%, 90.45% less memory consumptions, as well as 4.4 × , 5.0 × , 5.8 × , 6.1 × practical speedups on real hardware devices, while maintaining comparable accuracy. [ABSTRACT FROM AUTHOR]

Identifying standard fetal ultrasound (US) planes with key anatomical structures during mid-pregnancy prenatal screening is crucial for measuring fetal growth parameters and early detection of abnormalities. However, obtaining these standard planes is laborious and time-consuming and depends on the clinical experience of sonographers. Automatic detection of these planes can aid sonographers in identifying the correct standard planes. In recent times, various deep learning techniques have developed to automate the detection of standard fetal US planes. However, a common limitation among these approaches is their dependence on a single model prediction to make the final decision, which introduces the possibility of inaccuracies. Therefore, we propose an automated identification of commonly used standard fetal US planes based on the stacking ensemble of deep convolutional neural networks (CNN). The stacking ensemble method employs three pre-trained deep CNNs: AlexNet, VGG-19, and DarkNet-19. Softmax and random forest classifiers are used to get predictions from deep CNNs. The final prediction is made using the absolute majority voting technique. A publicly available fetal US dataset is employed to evaluate the performance of the stacking ensemble approach. The proposed ensemble model classifies fetal US planes into six distinct classes: abdomen, brain, femur, thorax, maternal cervix, and other (less commonly employed planes, such as kidney, and limbs) fetal planes. Experimental findings demonstrate that the stacking ensemble approach achieved high performance with an accuracy of 95.69%, precision of 94.02%, recall of 96.28%, F1-score of 95.08%, specificity of 99.12%, and Matthews correlation coefficient of 94.19% compared to individual deep CNN models and other competing methods. • This paper presents a stacking ensemble of CNNs to classify fetal US images. • The stacking ensemble learning model includes AlexNet, VGG-19, modified VGG-19 and DarkNet-19. • The final classification is made by using the absolute majority voting method. • A publicly available fetal US dataset is used to validate the proposed method. • Proposed method outperformed the state-of-the-art and competing algorithms. [ABSTRACT FROM AUTHOR]

In this paper, we introduce the Spatial Bias to learn global knowledge without self-attention in convolutional neural networks. Owing to the limited receptive field, conventional convolutional neural networks suffer from learning long-range dependencies. Non-local neural networks have struggled to learn global knowledge, but unavoidably have too heavy a network design due to the self-attention operation. Therefore, we propose a fast and lightweight Spatial Bias that efficiently encodes global knowledge without self-attention on convolutional neural networks. Spatial Bias is stacked on the feature map and convolved together to adjust the spatial structure of the convolutional features. Because we only use the convolution operation in this process, ours is lighter and faster than traditional methods based on the heavy self-attention operation. Therefore, we learn the global knowledge on the convolution layer directly with very few additional resources. Our method is very fast and lightweight due to the attention-free non-local method while improving the performance of neural networks considerably. Compared to non-local neural networks, the Spatial Bias use about × 10 times fewer parameters while achieving comparable performance with 1.6 ∼ 3.3 times more throughput on a very little budget. Furthermore, the Spatial Bias can be used with conventional non-local neural networks to further improve the performance of the backbone model. We show that the Spatial Bias achieves competitive performance that improves the classification accuracy by + 0. 79 % and + 1. 5 % on ImageNet-1K and CIFAR-100 datasets. Additionally, we validate our method on the MS-COCO and ADE20K datasets for downstream tasks involving object detection and semantic segmentation. • We introduce Spatial Bias, a lightweight non-local method in neural networks. • The proposed method learns long-range dependencies with a convolution layer. • The proposed Spatial Bias is fairly quick and show competitive performance. • The proposed Spatial Bias can be used with the existing non-local methods. [ABSTRACT FROM AUTHOR]

As the core for the lubrication and operation of space ball bearing, the porous non-metallic cage with oil-impregnated will cause catastrophic results in case of fault. The principal used reasons for not realizing its early fault diagnosis are: vibration, temperature, and sound signals may not carry the early fault information, and the traditional models could not work effectively in the case of insufficient data samples. Aiming at the above problems, this paper first presents a visual cage motion test rig, which realizes the accurate acquisition of cage dynamic signals without changing the bearing structure. The early fault dataset of the space ball bearing cage is then provided, composed of the data sample including the dynamic signal with early fault information of the cage. Finally, a Similarity Clustering Two-Channel Convolutional Neural Network (SC-TCCNN) model, which integrates the similarity clustering method into the Double-TCCNN framework with parameter sharing, is proposed. The experimental results show that the method proposed realizes the intelligent diagnosis of cage early fault and shows more competitiveness than the traditional model in the case of insufficient training data samples. [ABSTRACT FROM AUTHOR]

Medical image landmark detection can automatically determine the pre-set position coordinates in medical images, so as to further assist disease diagnosis and treatment planning. Using Convolutional Neural Network (CNN) to estimate the heatmap of the landmark has become the mainstream method. However, some landmarks are not obvious and there are local similarities between different landmarks in the medical images. Existing methods usually directly map the image features to all the landmarks, without considering these two challenges. In this paper, we propose a Feature Decouple and Gated Recalibration Network (FDGR-Net) for the medical image landmark detection. The FDGR-Net first establishes independent pathways to decouple all landmark features into single landmark features. In this process, a dynamic multi-kernel convolution is designed as the decouple basic component to improve the feature robustness of the not obvious landmarks. Then, a novel Gated Feature Recalibration Module (GFRM) is proposed, which includes a retention gate and an elimination gate. The retention gate can further strengthen the information of the target landmark and the elimination gate can eliminate the interference of non-target landmarks. The proposed method is validated on one public dataset and two in-house datasets, and its performance is superior to the existing advanced method. The Mean Radial Error (MRE) of the proposed method can reach 3.04, 5.51 and 1.45 in the Knee dataset, Hand dataset and Head dataset, respectively. • Feature decouple and gated recalibration network improves the medical image landmark detection. • A new dynamic multi-kernel convolution as decouple module's basic component to decouple features. • Gated feature recalibration module further eliminates the interference information of non-target landmarks. • The proposed method has excellent performance in landmark detection for a variety of medical images. [ABSTRACT FROM AUTHOR]

The Single-Line Diagram (SLD) is a standardized graphical representation of power systems. Power engineering software typically requires users to model power systems by drawing SLDs. Although existing computer vision-based SLD recognition approaches facilitate system modeling to some extent, their performance is usually limited by the low resolution of images, the imbalance in size and number of electrical symbols, and the interference between adjacent graphical elements on practical SLDs. This paper proposes a novel SLD recognition framework to address these challenges. First, a text and symbol separation module based on digital image processing is developed to enable independent analysis of text labels and electrical symbols to avoid mutual interference. To accurately recognize electrical symbols of different sizes, the proposed framework integrates the Swin Transformer into the Faster Region-based Convolutional Neural Network (Faster R-CNN), which ultimately achieves a minimum precision of 98.9% and a minimum recall of 98.12%. In addition, the framework modifies the Text Super-Resolution Network (TSRN) to improve text recognition performance on low-resolution images. Combined with the Attentional Scene Text Recognizer with Flexible Rectification (ASTER) model, the proposed text recognition model results in an average accuracy increase of 1% on the TextZoom dataset and 97.1% accuracy in recognizing practical SLD text labels. A synthetic SLD generation method is also proposed to augment the actual SLD datasets to overcome the class imbalance of electrical symbols. The proposed framework significantly improves the performance of existing engineering drawing recognition frameworks. Relevant source code is available online at https://github.com/Lattle-y/work.git. • An advanced framework for Single-Line Diagram element recognition is proposed. • An efficient graphic text separation algorithm is proposed. • Recognizing graphics and text using improved deep-learning algorithms. • Proposing a data generation method to cope with the lack of engineering drawings. • Experimental validation using real-world one-line diagrams provided by the electrical company. [ABSTRACT FROM AUTHOR]

Accurate and reliable load forecasting can ensure the safety and economy of power system operation. To improve the accuracy of short-term power load forecasting, this paper adopts feature extraction combined with a variety of machine learning methods,(Empirical wavelet decomposition (EWT), Convolutional neural network (CNN) model, Recurrent neural network (RNN) model, Long short-term memory network (LSTM) model, Bayesian optimization (BOA) algorithm). A new hybrid forecasting model, the EWT-CNN-S-RNN + LSTM model is proposed to predict power consumption. In this method, the fixed mode of EWT decomposition is used to extract statistical features, and the LSTM/RNN model is selected according to the statistical features. This method uses the Bayesian optimization (BOA) algorithm to optimize the parameters to solve the model gradient explosion problem. Based on the electricity consumption data of 2007 (Australian Energy Market Operator) and a total of 336 power load data, a good load forecasting result is obtained, which can provide indirect support for power regulation and sustainable development of electricity. [ABSTRACT FROM AUTHOR]

Motion capture (mocap) data stores the skeleton movement of recorded objects or humans and is essential for various 3D applications, such as games, animations, virtual reality, surveillance and medical. The exploration of steganography using mocap data is still limited despite its wide-range of potential practical applications. Compared to image, audio, and video steganography, the challenges in mocap steganography arise from differences between the temporal lengths and topologies of the cover and secret data, non-zero residuals between mocap data sequences, non-repetitive movements in different chunks of the mocap data. Deep neural network (DNN) based hiding models could be a potential solution to address these challenges in mocap steganography. DNN-based steganography has been adopted in many recent works and has shown promising results in steganographic techniques in various domains, including image, audio and video; however, to date, no prior work has investigated the application of DNN models for hiding mocap data. In this paper, we investigate solutions to deal with the challenges in mocap steganography and propose a novel end-to-end deep neural network model, DSteganoM, to hide a sequence of skeletal motions into another. The proposed model is capable of learning and exploring the redundancy of spatial and temporal Euclidean positional information in each skeletal motion for the hiding task. In addition, by introducing upsampling and pooling mechanisms in our network architecture, the DSteganoM model also can hide mocap data of different topologies with different dimensions. Experimental results show that the proposed DSteganoM model outperforms other state-of-the-art models and can visually reconstruct the cover and secret with high resistance to signal perturbation. Our work bridges the research gap in applying deep steganography to address the challenges in mocap data hiding. • A new deep neural network model for hiding mocap data inside another mocap data. • This model can hide mocap data of different topologies with different dimensions. • The secret mocap data can be reconstructed under noise corruption and perturbation. • Performance of the model is evaluated with five other state-of-the-art models. [ABSTRACT FROM AUTHOR]

Deep machine learning models including Convolutional Neural Networks (CNN) have been successful in the detection of Mild Cognitive Impairment (MCI) using medical images, questionnaires, and videos. This paper proposes a novel Multi-branch Classifier-Video Vision Transformer (MC-ViViT) model to distinguish MCI from those with normal cognition by analyzing facial features. The data comes from the I-CONECT, a behavioral intervention trial aimed at improving cognitive function by providing frequent video chats. MC-ViViT extracts spatiotemporal features of videos in one branch and augments representations by the MC module. The I-CONECT dataset is challenging as the dataset is imbalanced containing Hard-Easy and Positive–Negative samples, which impedes the performance of MC-ViViT. We propose a loss function for Hard-Easy and Positive–Negative Samples (HP Loss) by combining Focal loss and AD-CORRE loss to address the imbalanced problem. Our experimental results on the I-CONECT dataset show the great potential of MC-ViViT in predicting MCI with a high accuracy of 90.63% accuracy on some of the interview videos. • We propose MC-ViViT to detect MCI from facial videos in the I-CONECT Study. • We design the MC module to enrich the extracted spatio-temporal features. • We develop HP loss to address the inter-and intra-class imbalanced data. [ABSTRACT FROM AUTHOR]

Lung cancer is the leading cause of cancer death worldwide, and most patients are diagnosed with advanced stages for lack of symptoms in the early stages of the disease, leading to poor prognosis. Thus, it is an urgent need to propose an accurate and fast pulmonary nodule detection network. In this paper, a model fusion based 2.5D pulmonary nodule detection network is proposed. In the first stage, the Weight Box Fusion (WBF) algorithm is introduced to make a fusion of the nodule candidates proposed by two mighty detection networks and yield a more accurate bounding box of nodule candidates. This is to determine the adaptive rate of the false positive reduction stage. The downside of model fusion is that it would generate many false positive candidates. So in the second stage, an adaptive 3D Convolutional Neural Network (CNN) architecture is designed to reduce the false positive candidates. The architecture can adjust the weights of soft assignment according to the diameter of nodules. By the experimental results on publicly available LUNA16, it demonstrates that 97.4% sensitivity can be obtained with an average of only 9.67 candidates per scan in our work that can compete with the state-of-the-art (SOTA) works. Besides, a 0.965 competition performance metric (CPM) is obtained on LUN16 and SOTA results from 1/8 to 1 FPs/Scan, which greatly outperforms other existing methods. • A robust and novel computer-aided system for lung nodule detection is proposed. • The experimental result on the open dataset LUNA16 exceeds the state-of-the-art level. • The split-attention block is introduced to enhance the features of small nodules. • A nodule selection scheme is proposed which is valid in false positive reduction. • An adaptive and novel false positive reduction method is proposed. [ABSTRACT FROM AUTHOR]

To identify epilepsy, Electroencephalography (EEG) is an important and common tool used to study the electrical activity of the human brain. The machine learning-based classifier is utilized to detect the seizure by manually extracting the features from the EEG signals in previous works. Though, these effective benefits have been proved already with automatic feature extraction, they are unable to achieve the classification of multiple classes. Meanwhile, real-time epileptic seizure detection is unable to keep the capacity because the identifiable EGG is too long. Hence, this paper is decided to develop an enhanced deep learning architecture with EEG signal for performing automatic epileptic seizure detection. The EEG signal is collected from the standard datasets. The Fourier-Bessel Series Expansion-Based Empirical Wavelet Transform (FBSE-EWT) method is used to decompose the signal from the gathered data. In feature extraction, the decomposed signals are used and while extracting the signal features, the autoencoder is involved in that process. To reduce the computational overload, the relief-F feature ranking method is used for choosing the important signal features. The Long Short Term Memory (LSTM) and Multi-Scale Atrous-based Deep Convolutional Neural Networks (MSA-DCNN) are used that is named as Hybrid Deep Scheme (HDS) to detect epileptic seizures with the top-ranked features for epileptic seizure classification. The Black Widow Optimization (BWO) and the Spider Monkey Optimization (SMO) are combined to generate Adaptive Spider Monkey Black Widow Optimization (ASMBWO) that is employed to perform the parameter tuning into a classification technique. Using different measures, the experimental analysis is done between the conventional epileptic seizure detection and the proposed model to establish an enhanced performance of the proposed model. [ABSTRACT FROM AUTHOR]

Hardware Trojans (HT) are the most malicious components attacking modern integrated circuits (ICs). Information leakage, incorrect functionality, and overheating are the major effects of HT insertion. Due to the outsourcing of multi-stage digital circuit design and fabrication, HTs can be inserted by rogues at every stage. The types of HTs and insertion methods evolve continuously and their detection becomes more complex. Consequently, designing integrated circuits robust to HT insertion seems efficient. A prerequisite for such a strategy is the evaluation of integrated circuit vulnerability to HT insertion. There are several factors involved in the vulnerability of fabricated integrated circuits such as the distribution of white spaces in the IC layout, the ratio of unutilized routing resources, the amount of activity of the circuit's nodes, the geometrical properties of the circuit's gates, the delay of various paths, and etc. The interrelation of these diverse factors makes IC vulnerability modeling a challenging problem. In this paper, a deep neural network (DNN) based approach is developed to combine all the effective factors in IC vulnerability analysis. Firstly, we generate a comprehensive dataset containing various types and sizes of hardware Trojans inserted into benchmark circuits. The benchmark circuits used in this study are ISCAS 85, ISCAS 89, and ITC 99. Then using the transfer learning concept seven state-of-the-art deep convolutional neural networks are trained and verified using the constructed dataset to classify every integrated circuit into three vulnerability levels (low, moderate, and highly vulnerable classes). Simulation results show that our proposed method achieves almost 95 % accuracy in the most successful case (VGG16). The lowest accuracy belongs to Inception V3 (∼87.65 %) which is much better than the most successful previous studies (72 %). This is mainly because the DNN-based method handles shortcomings (such as incomplete modeling of vulnerability-related factors) in the previous state-of-the-art methods efficiently. [ABSTRACT FROM AUTHOR]

Knowledge distillation, a technique recently gaining popularity for enhancing model generalization in Convolutional Neural Networks (CNNs), operates under the assumption that both teacher and student models are trained on identical data distributions. However, its effect on Graph Neural Networks (GNNs) is less than satisfactory since the graph topology and node attributes are prone to evolve, thereby leading to the issue of distribution shift. In this paper, we tackle this challenge by simultaneously training a group of graph neural networks in an online distillation fashion, where the group knowledge plays a role as a dynamic virtual teacher and the structure changes in graph neural networks are effectively captured. To improve the distillation performance, two types of knowledge are transferred among the students to enhance each other: local knowledge reflecting information in the graph topology and node attributes, and global knowledge reflecting the prediction over classes. We transfer the global knowledge with KL-divergence as the vanilla knowledge distillation does, while exploiting the complicated structure of the local knowledge with an efficient adversarial cyclic learning framework. Extensive experiments verified the effectiveness of our proposed online adversarial distillation approach. The code is published at https://github.com/wangz3066/OnlineDistillGCN. • We propose an Online Adversarial Knowledge Distillation framework for GNNs. • OAD trains a group of GNN models to capture structure updates in GNNs. • OAD extracts both local and global knowledge in graph structured data. • Extensive experiments demonstrate the effectiveness of OAD. [ABSTRACT FROM AUTHOR]

Accurately estimating object numbers plays a crucial role in industrial applications. However, object counting in computer vision poses a formidable challenge, particularly when dealing with images containing tiny, similar, and stacked objects. Therefore, this paper proposes an encoder–decoder-based convolutional neural network named AMSA-CAFF Net to accurately count and regress high-quality density maps from X-ray images of highly dense microscopic components. AMSA-CAFF Net has three major components: an adaptive multi-scale feature aggregation (AMSA) module for encoding adaptive multi-scale features in the encoder, a channel-wise adaptive feature fusion (CAFF) module to ensure smooth feature fusion, and a continuous feature enhancement mechanism to promote information integration in the decoder. In addition, there is still a lack of datasets dedicated to industrial object counting tasks. To address the limitations associated with the existing datasets, we construct a novel and large-scale object counting dataset named X-ray-based industrial electronic component counting dataset (XRAY-IECCD), consisting of 1,460 X-ray images of 10 types of components, resulting in a total of 2,915,126 objects annotated with points. To the best of our knowledge, the XRAY-IECCD is the first dataset of tiny, dense, multi-class object counts with a substantial number of annotations. We evaluate the performance of AMSA-CAFF Net on three datasets: the XRAY-IECCD, crowd counting benchmark ShanghaiTech dataset, and object counting benchmark CARPK dataset. The proposed framework outperforms the state-of-the-art approaches over the proposed XRAY-IECCD and CARPK datasets, highlighting its effectiveness. Furthermore, we achieve competitive results on the ShanghaiTech dataset, which demonstrates the generalization capability of our method. [ABSTRACT FROM AUTHOR]

The stable control of product quality when abnormal working conditions occur in the industrial production process is essential to improve product quality and economic efficiency. However, the process industry suffers from multivariate, nonlinear, uncertainty, long time delays and frequent failures, making its modeling and control difficult. In this paper, a novel method of neural network model predictive control integrated process monitoring (PM-NNMPC) is proposed. First, the combination of gated recurrent unit and convolutional neural network (GRU-CNN) is used to extract the features from the time and the space separately to build prediction models for different working conditions of the process. Then, a process monitoring and fault diagnosis method of principal component analysis combined with deep neural network and XGBoost (PCA-DNN-XGBoost) is designed to monitor the working conditions in real-time and diagnose accurately when faults occur. Finally, a new operation control framework integrating process monitoring and fault diagnosis into the model prediction control is designed to solve the problem that equipment operation failure, which makes the product quality fluctuation because of the model mismatch and the system runaway. The experimental results of hot rolling process show that PM-NNMPC can effectively control the system under normal working conditions and in case of the system failure to ensure the stable product quality. [ABSTRACT FROM AUTHOR]

Sputum deposition has always been a significant problem in patients with mechanical ventilation. If not handled in time, it can induce bacterial infection and even endanger the life safety of patients. Currently, the evaluation of sputum deposition heavily depends on the medical staff's clinical experience. This paper designs a sputum deposition monitoring system that realizes remote airway pressure and flow signals collection and management for mechanically ventilated patients. Forty-six patients in the intensive care unit were involved in this study. Meanwhile, a one-dimensional convolution neural networks model was proposed to classify four sputum deposition categories (no, slight, moderate, and severe). The experimental results showed that the overall classification accuracy could reach more than 78%. Moreover, the model has been optimized for practical application by setting thresholds for the output of the softmax layer. Finally, the classification accuracy of no sputum, slight, moderate, and severe deposition reaches 85.84%, 84.29%, 93.19%, and 93.38%, respectively. This study's proposed system and method could significantly increase the automation and intelligence of medical care. [ABSTRACT FROM AUTHOR]

The presence of Beggiatoa Spp. indicates anoxic conditions or 'poor condition' in marine sediments beneath aquaculture pens, resulting from organic enrichment. Currently, the most efficient approach to estimate Beggiatoa Spp. coverage, and thus the extent of the issue, involves video surveys which are scored by human observers for presence of this bacteria. However, this approach is highly time-consuming and relies heavily on the expertise and experience of the individuals involved, thus affecting its accuracy. Machine learning-based computer vision techniques, such as Convolutional Neural Networks (CNNs), offer the potential for automated estimation of Beggiatoa Spp. coverage. However, most existing machine learning methods focus solely on the estimation of the coverage via presence, absence of a single type of Beggiatoa Spp.. These approaches typically rely on binary classification to distinguish the object from the background when estimating coverage. Nevertheless, the inclusion of subordinate categories within high-level classifications poses a great challenge for accurately estimating their coverage rates. In this paper, an adaptive ensemble learning approach was proposed to estimate Beggiatoa Spp. coverage. Unlike other approaches, the proposed approach is capable in adaptively extracting and fusing features from underwater images and accurately estimating the coverages of multiple types of Beggiatoa Spp. through ensemble learning. Experimental results demonstrated that our proposed approach outperforms other approaches in terms of both effectiveness and efficiency in estimating Beggiatoa Spp. coverage. • Adaptive feature extraction and fusion enhance relevant and informative features' identification. • Adaptive feature extraction and fusion enhance feature representation and reduce time cost. • Flexible ensemble learning framework improves accuracy, robustness, and efficiency. • Flexible ensemble learning framework can be extended to other models or applications. • Novel machine learning-based approach estimates subordinate categories of Beggiatoa Spp. coverage. [ABSTRACT FROM AUTHOR]

Recent studies focus on the utilization of deep learning approaches to recognize depression from facial videos. However, these approaches have been hindered by their limited performance, which can be attributed to the inadequate consideration of global spatial–temporal relationships in significant local regions within faces. In this paper, we propose Spatial–Temporal Attention Depression Recognition Network (STA-DRN) for depression recognition to enhance feature extraction and increase the relevance of depression recognition by capturing the global and local spatial–temporal information. Our proposed approach includes a novel Spatial–Temporal Attention (STA) mechanism, which generates spatial and temporal attention vectors to capture the global and local spatial–temporal relationships of features. To the best of our knowledge, this is the first attempt to incorporate pixel-wise STA mechanisms for depression recognition based on 3D video analysis. Additionally, we propose an attention vector-wise fusion strategy in the STA module, which combines information from both spatial and temporal domains. We then design the STA-DRN by stacking STA modules ResNet-style. The experimental results on AVEC 2013 and AVEC 2014 show that our method achieves competitive performance, with mean absolute error/root mean square error (MAE/RMSE) scores of 6.15/7.98 and 6.00/7.75, respectively. Moreover, visualization analysis demonstrates that the STA-DRN responds significantly in specific locations related to depression. The code is available at: https://github.com/divertingPan/STA-DRN. [ABSTRACT FROM AUTHOR]

Use of social media for communication, sharing or expressing views, broadcasting news, threatening and blackmailing has become an integral part of society. One such activity is understanding multi-cultural wedding images uploaded on social media. This paper presents a novel method based on the combination of U-Net, Convolutional Neural Network and Random Forest for classification of multicultural wedding images. In the case of wedding images, bride and bridegroom draw the attention of the viewers. This observation led to propose a U-Net for segmenting the region of bride and bridegroom in a novel way. Similarly, it is noted that the costumes of bride and bridegroom are vital information for differentiating different cultures. This cue motivated us to extract features using CNN for classification. Since the extracted features using CNN are capable of discriminating images of different classes, we propose a simple and effective Random-Forest for Multicultural Wedding Image Classification. The efficiency of the proposed model is demonstrated by testing it on our own dataset of six multi-cultural wedding classes and standard dataset of wedding and non-wedding images classes. Experimental results on both the datasets show that the proposed model outperforms the state-of-the-art models in terms of average classification rate. [ABSTRACT FROM AUTHOR]

Ultrasound assessment of fetal biometry is widely used for gestational age prediction and abnormal fetal growth diagnosis. However, manual biometric measurements require time-consuming procedures and current automated measurement approaches remain challenging in terms of accuracy and computational complexity. This paper proposes deep learning-based efficient automatic fetal biometry measurement method for system-on-chip (SoC) solution. For end-to-end automated measurements, a hardware-friendly bilateral segmentation network (H-BiSeNet) was designed and optimized for multiple fetal objects (i.e., head, abdomen and femur), and fetal biometric parameters were then calculated with a robust multiple keypoint detection algorithm. For hardware implementation with a SoC, a deep learning processing unit (DPU) was employed to accelerate the proposed segmentation network, and biometric measurements with image pre- and postprocessing were conducted on an application processing unit (APU). A total of 6000 fetal ultrasound images in the three regions were collected from 2568 subjects, and the dataset was utilized to train the proposed network. In the performance evaluation, the proposed segmentation network with optimal quantization (H-BiSeNet-Q) outperformed other segmentation networks (i.e., U-Net, U-Net++, Attention U-Net, U-Net3+, DeepLabv3 + and BiSeNetv2) in terms of the average Dice coefficient, and automated measurements with H-BiSeNet-Q were highly correlated with manual measurements (r ≥ 0.983, p < 0.01 for all parameters, i.e., biparietal diameter, head circumference, abdominal circumference and femur length). In the execution time analysis, H-BiSeNet-Q ran at 4.25 frames per second (FPS) with parallelism corresponding to 80% resource utilization in the SoC. These results showed feasibility in low-resource hardware settings such as portable ultrasound systems. [ABSTRACT FROM AUTHOR]

In recent studies, Transformer collaborated with convolution neural network (CNN) have made certain progress in the field of intelligent fault diagnosis by leveraging their respective advantages of global and local feature extraction. However, the multihead self-attention block used by Transformer and cross-channel convolution mechanism existing in CNN would make the collaborative models overly complex, leading to higher hardware requirements and limited industrial application scenarios. Therefore, this paper proposes a lightweight fault diagnosis framework called LiConvFormer to address the aforementioned challenges. First, a separable multiscale convolution block is designed to extract multilocal receptive field features of vibration signals and greatly reduce the learning parameters and computations. Second, a broadcast self-attention block is developed to capture critical fine-grained features within the signal's global scope, while avoiding cumbersome operations such as matrix multiplication and multidimensional exponentiation. Experimental results on three mechanical systems show that the proposed framework can accommodate advantages of lightweight and robustness compared to the recent Transformer and CNN-based fault diagnosis methods; moreover, the superiority of the above two blocks is also verified. The code library is available at: https://github.com/yanshen0210/LiConvFormer-a-lightweight-fault-diagnosis-framework. [ABSTRACT FROM AUTHOR]

A technique of analysis of thermal images of minigrinders is presented in the paper. The following states of minigrinders were analyzed: healthy minigrinder (HM), minigrinder with 2 blocked air vents (M2BAV), and minigrinder with 4 blocked air vents (M4BAV). The thermal imaging camera was shivering vertically (0–1.5 [m/s2]). Two methods are proposed for the computation of essential areas of thermal images: DTILE (Differences of Thermal Images using Local Entropy), and DPTILE (Differences of Power of Thermal Images using Local Entropy). The computed areas are compared to each other using the Nearest Neighbor. Moreover, verification of the analysis of thermal images was carried out for the convolutional neural networks GoogLeNet and ResNet-50. The analyzed approaches were proper for fault diagnosis. The efficiency of recognition is equal to 100% for the three analyzed classes. The presented approach can be also used in the industry. [ABSTRACT FROM AUTHOR]

Despite the high efficiency of keyhole plasma arc welding (K-PAW), it still has several deficiencies, such as narrow welding parameter ranges, easily disturbed welding process and instability of welding quality, etc. It is a significant prerequisite to predict the keyhole/penetration status accurately for maintaining the welding process stability and improving the welding quality. Most researchers have focused on visual inspection techniques and convolutional neural networks (CNN), establishing a mathematical model of the correlation between weld pool images and penetration status. While CNN could extract the features of single weld pool images, it is difficult to predict the evolution trend of the incoming moments. In this paper, a novel model based on CNN and LSTM (long-short term memory) was developed to extract both the spatial and temporal features of topside weld pool images, and consequently the complex keyhole behaviors were predicted and described. The comparative study is carried out on different models, i.e. the single CNN model, the single LSTM model, and the CNN-LSTM model that integrates both spatial features of single images and temporal features of sequence. The keyhole initialization and establishing period, as the typical and critical welding scenario in full-penetration welding, is investigate and compared, resulting in a predicted value that is close to reality. Furthermore, ahead prediction of keyhole status was adopted, maintaining over 80% accuracy even when predicting keyhole behaviors 2 s into the future. Consequently, the unified CNN-LSTM model effectively improves the prediction accuracy of keyhole/penetration status, promising for intelligent K-PAW technology. [ABSTRACT FROM AUTHOR]

In the context of a resilient energy system, accurate residential load forecasting has become a non-trivial requirement for ensuring effective management and planning strategy/policy development. Due to the highly stochastic nature of energy load profiles, it is difficult to predict accurately, and usually, predictions are error-prone. This paper explores the potential of Empirical Mode Decomposition (EMD) in simplifying the dynamics of complex demand profiles. The simplified components are then embedded within a deep learning model, specifically Convolution Neural Network (CNN) and Long Short-Term Memory (LSTM), to forecast short-term residential loads. The novel modelling framework integrates Bayesian optimisation strategy, feature decomposition technique, feature engineering phase, and percentile-based bias correction algorithm to enhance model accuracy. The model is developed using a case-study residential dwelling located in Fintry (Scotland), and the model performance is assessed over four forecast horizons. The overall efficiency of framework is also investigated for three algorithms: random forest, gradient boosting decision trees (GBDT), and an LSTM network. While EMD and feature engineering were found to greatly improve prediction accuracy, the number of IMFs used was shown to significantly impact the model's performance and computational complexity. The model was tested on two further case studies from Fintry. • A novel forecasting framework is proposed for short-term energy demand prediction. • Multi-domain features are extracted to increase the accuracy. • Utilising empirical mode decomposition to enhance framework performance. • Exhibit strong forecasting capability without the additional computational cost. • Excellent forecasting accuracy and model robustness are validated using three households. [ABSTRACT FROM AUTHOR]

The Computational Fluid Dynamics (CFD) simulations at different fidelity levels are a common tool for predicting the aerodynamic performance in many engineering applications such as; automotive and aerospace to cite a few. However, the associated computational costs with these simulations are not trivial, and using the CFD simulations might easily become unpractical specially at the early design stages. The number of the required CFD simulations could be reduced immensely by using Artificial Intelligence (AI) methods which can predict, for example, the aerodynamic coefficients based on previous and limited CFD simulations. Having discussed the AI potential to minimize the CFD usage, the objective of this work is to investigate and compare the ability of different AI methods to predict the aerodynamic coefficients using the same CFD database from the literature. In this study, three different AI methods, namely; machine learning (e.g. Extreme Gradient Boosting (XGB), CatBoost, Bagging, Light Gradient Boosting Machine (Light-GBM), Random Forest (RFR), Gradient Boosting), deep learning (e.g. One Dimensional Convolutional Neural Network (Conv1D)) and surrogate models (e.g. Gaussian Process Regression (GPR)) have been utilized to estimate a sample CFD database (e.g. NACA0012, RAE2822) from the existing literature, and the prediction performance of these methods have been compared. The results show that, the aerodynamic coefficients predicted using CatBoost, XGB and Bagging are in good agreement with the reference CFD results. Furthermore, the proposed Conv1D method with LeakyReLU activation function showed a promising results in the development of aerodynamic models which can be used in the early stages of the air vehicles design. The obtained results from this work indicate that the CatBoost and the XGB models required much smaller training time compared to the Conv1D method, moreover, the proposed regression trees; CatBoost, Bagging and XGB can reduce the required number of CFD simulations significantly. • Various artificial intelligence methods were studied to predict aerodynamic data. • Regression tree models such as XGB, Catboost and LightGBM were used. • Surrogate models and deep learning models (i.e. Conv1D) also were used. • Detailed comparison were presented between the aforementioned models. • The paper suggests the suitability of XGB and Catboost for aerodynamics cases. [ABSTRACT FROM AUTHOR]

Nowadays, users across the globe interact with one another for information exchange, communication, and association on various online social media. However, some individuals exploit these venues for malicious practices like hate speech and cyberbully. In this paper, we present an improved multilingual hate speech and cyberbully detection model using bagging-stacking based hybrid ensemble deep learning techniques. The proposed model utilizes Bi-directional Long Short-Term Memory (BiLSTM), Bi-directional Gated Recurrent Unit (Bi-GRU), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM) techniques to enhance the overall performance. We first preprocess the multilingual data streams followed by adoption of Global vectors for word Representation (GloVe) embeddings to convert words to a vector representation in parallel enabling the data streams for binary classification task. In order to construct an architecture for the detection of hate speech and cyberbully, we introduce a heterogeneous fusion of multiple effective models in a unique approach such that CNN-LSTM utilizes a stacking approach with stochastic gradient descent to achieve optimal weights, whereas all the base learners used bagging ensemble approach with cross-validation to reach optimal weights. The final output layer of the proposed ensemble deep learning architecture is achieved using a super learner approach on base learners. To show the efficacy of the proposed model, we conduct the simulation on a total of nine real-world social media datasets in different languages and compared the results with other contemporary hate speech and cyberbully detection methods. The collected findings show that the proposed model outperforms other models on considered datasets and shows an improvement of at least 4.44% in F1 scores. • Proposed an ensemble DL multilingual model for hate speech & cyberbully detection. • Model combines bagging, stacking, and super learning methods to improved results. • Model adopts a light weight heterogeneous fusion and bagging-stacking hybrid methods. • The experimental findings advocate the superiority of the proposed model. [ABSTRACT FROM AUTHOR]