Search

Abstract Radar observation variables reflect the precipitation amount of strong convective precipitation processes, which accurate forecast is an important difficulty in weather forecasting. Current forecasting methods are mostly based on radar echo extrapolation, which has the insufficiency of input information and the ineffectiveness of model architecture. This paper presents a Bidirectional Long Short-Term Memory forecasting method for strong convective precipitation based on the attention mechanism and residual neural network (ResNet-Attention-BiLSTM). First, this paper uses ResNet to effectively extract the key information of extreme weather and solves the problem of regression to the mean of the prediction model by learning the residuals of the radar observation data. Second, this paper uses the attention mechanism to adaptively weight the fusion of the features to enhance the extraction of the important features of the precipitation image data. On this basis, this paper presents a novel spatio-temporal reasoning method for radar observations and establishes a precipitation forecasting model, which captures the past and future time-order relationship of the sequence data. Finally, this paper conducts experiments based on the real collected data of a strong convective precipitation process and compares its performance with the existing models, the mean absolute percentage error of this model was reduced by 15.94% (1 km), 18.72% (3 km), and 14.91% (7 km), and the coefficient of determination ( $$R^{2}$$ R 2 ) was increased by 10.89% (1 km), 9.61% (3 km), and 9.29% (7 km), which proves the state of the art and effectiveness of this forecasting model.

Abstract Building extraction aims to extract building pixels from remote sensing imagery, which plays a significant role in urban planning, dynamic urban monitoring, and many other applications. UNet3+ is widely applied in building extraction from remote sensing images. However, it still faces issues such as low segmentation accuracy, imprecise boundary delineation, and the complexity of network models. Therefore, based on the UNet3+ model, this paper proposes a 3D Joint Attention (3DJA) module that effectively enhances the correlation between local and global features, obtaining more accurate object semantic information and enhancing feature representation. The 3DJA module models semantic interdependence in the vertical and horizontal dimensions to obtain feature map spatial encoding information, as well as in the channel dimensions to increase the correlation between dependent channel graphs. In addition, a bottleneck module is constructed to reduce the number of network parameters and improve model training efficiency. Many experiments are conducted on publicly accessible WHU,INRIA and Massachusetts building dataset, and the benchmarks, BOMSC-Net, CVNet, SCA-Net, SPCL-Net, ACMFNet, MFCF-Net models are selected for comparison with the 3DJA-UNet3+ model proposed in this paper. The experimental results show that 3DJA-UNet3+ achieves competitive results in three evaluation indicators: overall accuracy, mean intersection over union, and F1-score. The code will be available at https://github.com/EnjiLi/3DJA-UNet3Plus .

Abstract In transportation, roads sometimes have cracks due to overloading and other reasons, which seriously affect driving safety, and it is crucial to identify and fill road cracks in time. Aiming at the defects of existing semantic segmentation models that have degraded the segmentation performance of road crack images and the standard convolution makes it challenging to capture the spatial and channel coupling relationship between pixels. It is difficult to differentiate crack pixels from background pixels in complex backgrounds; this paper proposes a semantic segmentation model for road cracks that combines channel-spatial convolution with the aggregation of frequency features. A new convolutional block is proposed to accurately identify cracked pixels by grouping spatial displacements and convolutional kernel weight dynamization while modeling pixel spatial relationships linked to channel features. To enhance the contrast of crack edges, a frequency domain feature aggregation module is proposed, which uses a simple windowing strategy to solve the problem of mismatch of frequency domain inputs and, at the same time, takes into account the effect of the frequency imaginary part on the features to model the deep frequency features effectively. Finally, a feature refinement module is designed to refine the semantic features to improve the segmentation accuracy. Many experiments have proved that the model proposed in this paper has better performance and more application potential than the current popular general model.

Abstract In the paper, a new evolutionary technique called Linear Matrix Genetic Programming (LMGP) is proposed. It is a matrix extension of Linear Genetic Programming and its application is data-driven black-box control-oriented modeling in conditions of limited access to training data. In LMGP, the model is in the form of an evolutionarily-shaped program which is a sequence of matrix operations. Since the program has a hidden state, running it for a sequence of input data has a similar effect to using well-known recurrent neural networks such as Long Short-Term Memory (LSTM) or Gated Recurrent Unit (GRU). To verify the effectiveness of the LMGP, it was compared with different types of neural networks. The task of all the compared techniques was to reproduce the behavior of a nonlinear model of an underwater vehicle. The results of the comparative tests are reported in the paper and they show that the LMGP can quickly find an effective and very simple solution to the given problem. Moreover, a detailed comparison of models, generated by LMGP and LSTM/GRU, revealed that the former are up to four times more accurate than the latter in reproducing vehicle behavior.

Abstract This paper presents a comparison of machine learning (ML) methods used for three-dimensional localization of partial discharges (PD) in a power transformer tank. The study examines ML and deep learning (DL) methods, ranging from support vector machines (SVM) to more complex approaches like convolutional neural networks (CNN). Multiple case studies are considered, each with different attributes, including sensor position, frequency content of the PD signal, and size of the transformer tank. The paper focuses on predicting the PD location in three-dimensional space using single-sensor electric field measurements. Various aspects of each method are analyzed, such as the input signal, core methodology, correlation coefficient between the predicted location and the actual location, and root mean square error (RMSE). These features are discussed and compared across the different methods. The results indicate that the CNN model exhibits superior performance in terms of location accuracy among the methods considered.

Abstract With a significant number of mechanical products approaching the retirement phase, the batch recycling of discarded mechanical parts necessitates a preliminary assessment of their surface condition. However, the presence of surface rust poses a challenge to defect identification. Therefore, this paper proposes a method for detecting heavily rusted surface defects based on an improved YOLOv8n network. In the Backbone, the C2f-DBB module of re-parameterized deep feature extraction was introduced, and the attention module was designed to improve the accuracy of information extraction. In the Neck part, a Bi-Afpn multiscale feature fusion strategy is designed to facilitate information exchange between features at different scales. Finally, Focal-CIoU is employed as the bounding box loss function to enhance the network’s localization performance and accuracy for defects. Experimentally, it is proved that the improved network in this paper improves the Recall, Precision, and mAP0.5 by 1.2%, 2.1%, and 1.9%, respectively, on the original basis, which is better than other network models.

Abstract The vigorous development of the construction industry has also brought unprecedented safety risks. The wearing of safety helmets at the construction site can effectively reduce casualties. As a result, this paper suggests employing a deep learning-based approach for the real-time detection of safety helmet usage among construction workers. Based on the selected YOLOv5s network through experiments, this paper analyzes its training results. Considering its poor detection effect on small objects and occluded objects. Therefore, multiple attention mechanisms are used to improve the YOLOv5s network, the feature pyramid network is improved into a BiFPN bidirectional feature pyramid network, and the post-processing method NMS is improved into Soft-NMS. Based on the above-improved method, the loss function is improved to enhance the convergence speed of the model and improve the detection speed. We propose a network model called BiFEL-YOLOv5s, which combines the BiFPN network and Focal-EIoU Loss to improve YOLOv5s. The average precision of the model is increased by 0.9% the recall rate is increased by 2.8%, and the detection speed of the model does not decrease too much. It is better suited for real-time safety helmet object detection, addressing the requirements of helmet detection across various work scenarios.

Abstract In the field of spinal pathology, sagittal balance of the spine is usually judged by the spatial structure and morphology of pelvis, which can be represented by pelvic parameters. Pelvic parameters, including pelvic incidence, pelvic tilt and sacral slope, are therefore essential for the diagnosis and treatment of spinal disorders, however, it is a time-consuming and laborious procedure to measure these parameters by traditional methods. In this paper, an automatic measurement framework for pelvic CT images was proposed to calculate three-dimensional (3D) pelvic parameters with the support of deep learning technology. Pelvic images were first preprocessed, and 3D reconstruction was then performed to obtain 3D pelvic model by the Visualization Toolkit. DRINet was trained to segment the femoral head region in the pelvic images, and 3D sphere fitting was performed to locate the femoral heads. In addition, VGG16 was adopted to recognize images containing superior sacral endplate, and the plane growth algorithm was used to fit the plane so that the midpoint and normal vector of the superior sacral endplate could be obtained. Finally, 3D pelvic parameters were automatically calculated, and compared with manual measurements for 15 patients. The proposed framework automatically generated 3D pelvic models, and calculated two-dimensional (2D) and 3D pelvic parameters from continuous CT images. Experiments demonstrated that the framework can greatly speed up the calculation of pelvic parameters, and these parameters are accurate when compared with the manual measurements. In conclusion, the proposed framework demonstrates good performance on automatic pelvimetry measurement by incorporating deep learning technology, and can well replace the traditional methods for pelvic parameter measurement.

Abstract Across the globe, many transport bodies are advocating for increased cycling due to its health and environmental benefits. Yet, the real and perceived dangers of urban cycling remain obstacles. While serious injuries and fatalities in cycling are infrequent, “near misses”-events where a person on a bike is forced to avoid a potential crash or is unsettled by a close vehicle-are more prevalent. To understand these occurrences, researchers have turned to naturalistic studies, attaching various sensors like video cameras to bikes or cyclists. This sensor data holds the potential to unravel the risks cyclists face. Still, the sheer amount of video data often demands manual processing, limiting the scope of such studies. In this paper, we unveil a cutting-edge computer vision framework tailored for automated near-miss video analysis and for detecting various associated risk factors. Additionally, the framework can understand the statistical significance of various risk factors, providing a comprehensive understanding of the issues faced by cyclists. We shed light on the pronounced effects of factors like glare, vehicle and pedestrian presence, examining their roles in near misses through Granger causality with varied time lags. This framework enables the automated detection of multiple factors and understanding their significant weight, thus enhancing the efficiency and scope of naturalistic cycling studies. As future work, this research opens the possibility of integrating this AI framework into edge sensors through embedded AI, enabling real-time analysis.

Abstract To ensure the reliability of machining quality, it is crucial to predict tool wear accurately. In this paper, a novel deep learning-based model is proposed, which synthesizes the advantages of power spectral density (PSD), convolutional neural networks (CNN), and vision transformer model (ViT), namely PSD-CVT. PSD maps can provide a comprehensive understanding of the spectral characteristics of the signals. It makes the spectral characteristics more obvious and makes it easy to analyze and compare different signals. CNN focuses on local feature extraction, which can capture local information such as the texture, edge, and shape of the image, while the attention mechanism in ViT can effectively capture the global structure and long-range dependencies present in the image. Two fully connected layers with a ReLU function are used to obtain the predicted tool wear values. The experimental results on the PHM 2010 dataset demonstrate that the proposed model has higher prediction accuracy than the CNN model or ViT model alone, as well as outperforms several existing methods in accurately predicting tool wear. The proposed prediction method can also be applied to predict tool wear in other machining fields.

Abstract Automated defect detection in electroluminescence (EL) images of photovoltaic (PV) modules on production lines remains a significant challenge, crucial for replacing labor-intensive and costly manual inspections and enhancing production capacity. This paper presents a novel PV defect detection algorithm that leverages the YOLO architecture, integrating an attention mechanism and the Transformer module. We introduce a polarized self-attention mechanism in the feature extraction stage, enabling separate extraction of spatial and semantic features of PV modules, combined with the original input features, to enhance the network’s feature representation capabilities. Subsequently, we integrate the proposed CNN Combined Transformer (CCT) module into the model. The CCT module employs the transformer to extract contextual semantic information more effectively, improving detection accuracy. The experimental results demonstrate that the proposed method achieves a 77.9% mAP50 on the PVEL-AD dataset while preserving real-time detection capabilities. This method enhances the mAP50 by 17.2% compared to the baseline, and the mAP50:95 metric exhibits an 8.4% increase over the baseline.

Abstract COVID-19 has resulted in a significant global impact on health, the economy, education, and daily life. The disease can range from mild to severe, with individuals over 65 or those with underlying medical conditions being more susceptible to severe illness. Early testing and isolation are vital due to the virus’s variable incubation period. Chest radiographs (CXR) have gained importance as a diagnostic tool due to their efficiency and reduced radiation exposure compared to CT scans. However, the sensitivity of CXR in detecting COVID-19 may be lower. This paper introduces a deep learning framework for accurate COVID-19 classification and severity prediction using CXR images. U-Net is used for lung segmentation, achieving a precision of 0.9924. Classification is performed using a Convulation-capsule network, with high true positive rates of 86% for COVID-19, 93% for pneumonia, and 85% for normal cases. Severity assessment employs ResNet50, VGG-16, and DenseNet201, with DenseNet201 showing superior accuracy. Empirical results, validated with 95% confidence intervals, confirm the framework’s reliability and robustness. This integration of advanced deep learning techniques with radiological imaging enhances early detection and severity assessment, improving patient management and resource allocation in clinical settings.

Abstract Computer-aided diagnosis has been slow to develop in the field of oral ulcers. One of the major reasons for this is the lack of publicly available datasets. However, oral ulcers have cancerous lesions and their mortality rate is high. The ability to recognize oral ulcers at an early stage in a timely and effective manner is a very critical issue. In recent years, although there exists a small group of researchers working on these, the datasets are private. Therefore to address this challenge, in this paper a multi-tasking oral ulcer dataset (Autooral) containing two major tasks of lesion segmentation and classification is proposed and made publicly available. To the best of our knowledge, we are the first team to make publicly available an oral ulcer dataset with multi-tasking. In addition, we propose a novel modeling framework, HF-UNet, for segmenting oral ulcer lesion regions. Specifically, the proposed high-order focus interaction module (HFblock) performs acquisition of global properties and focus for acquisition of local properties through high-order attention. The proposed lesion localization module (LL-M) employs a novel hybrid sobel filter, which improves the recognition of ulcer edges. Experimental results on the proposed Autooral dataset show that our proposed HF-UNet segmentation of oral ulcers achieves a DSC value of about 0.80 and the inference memory occupies only 2029 MB. The proposed method guarantees a low running load while maintaining a high-performance segmentation capability. The proposed Autooral dataset and code are available from https://github.com/wurenkai/HF-UNet-and-Autooral-dataset .

Abstract Transcranial magnetic stimulation (TMS) has emerged as a promising neuromodulation technique with both therapeutic and diagnostic applications. As accurate coil placement is known to be essential for focal stimulation, computational models have been established to help find the optimal coil positioning by maximizing electric fields at the cortical target. While these numerical simulations provide realistic and subject-specific field distributions, they are computationally demanding, precluding their use in real-time applications. In this paper, we developed a novel multi-task deep neural network which simultaneously predicts the optimal coil placement for a given cortical target as well as the associated TMS-induced electric field. Trained on large amounts of preceding numerical optimizations, the Attention U-Net-based neural surrogate provided accurate coil optimizations in only 35 ms, a fraction of time compared to the state-of-the-art numerical framework. The mean errors on the position estimates were below 2 mm, i.e., smaller than previously reported manual coil positioning errors. The predicted electric fields were also highly correlated (r> 0.97) with their numerical references. In addition to healthy subjects, we validated our approach also in glioblastoma patients. We first statistically underlined the importance of using realistic heterogeneous tumor conductivities instead of simply adopting values from the surrounding healthy tissue. Second, applying the trained neural surrogate to tumor patients yielded similar accurate positioning and electric field estimates as in healthy subjects. Our findings provide a promising framework for future real-time electric field-optimized TMS applications.

Abstract As computer image processing and digital technologies advance, creating an efficient method for classifying sports images is crucial for the rapid retrieval and management of large image datasets. Traditional manual methods for classifying sports images are impractical for large-scale data and often inaccurate when distinguishing similar images. This paper introduces an SE module that adaptively adjusts the weights of input feature mapping channels, and a Res module that excels in deep feature extraction, preventing gradient vanishing, multi-scale processing, and enhancing generalization in image recognition. Through extensive experimentation on network structure adjustments, the SE-RES-CNN neural network model is applied to sports image classification. The model is trained on a sports image classification dataset from Kaggle, alongside VGG-16 and ResNet50 models. Training results show that the proposed SE-RES-CNN model improves classification accuracy by approximately 5% compared to VGG-16 and ResNet50 models. Testing revealed that the SE-RES-CNN model classifies 100 out of 500 sports images in 6 s, achieving an accuracy rate of up to 98% and a single prediction time of 0.012 s. This validates the model's accuracy and effectiveness, significantly enhancing sports image retrieval and classification efficiency. This validates the model's accuracy and effectiveness, significantly enhancing sports image retrieval and classification efficiency.

Abstract The effective detection of safflower in the field is crucial for implementing automated visual navigation and harvesting systems. Due to the small physical size of safflower clusters, their dense spatial distribution, and the complexity of field scenes, current target detection technologies face several challenges in safflower detection, such as insufficient accuracy and high computational demands. Therefore, this paper introduces an improved safflower target detection model based on YOLOv5, termed Safflower-YOLO (SF-YOLO). This model employs Ghost_conv to replace traditional convolution blocks in the backbone network, significantly enhancing computational efficiency. Furthermore, the CBAM attention mechanism is integrated into the backbone network, and a combined $$L_{CIOU + NWD}$$ L C I O U + N W D loss function is introduced to allow for more precise feature extraction, enhanced adaptive fusion capabilities, and accelerated loss convergence. Anchor boxes, updated through K-means clustering, are used to replace the original anchors, enabling the model to better adapt to the multi-scale information of safflowers in the field. Data augmentation techniques such as Gaussian blur, noise addition, sharpening, and channel shuffling are applied to the dataset to maintain robustness against variations in lighting, noise, and visual angles. Experimental results demonstrate that SF-YOLO surpasses the original YOLOv5s model, with reductions in GFlops and Params from 15.8 to 13.2 G and 7.013 to 5.34 M, respectively, representing decreases of 16.6% and 23.9%. Concurrently, SF-YOLO’s mAP0.5 increases by 1.3%, reaching 95.3%. This work enhances the accuracy of safflower detection in complex agricultural environments, providing a reference for subsequent autonomous visual navigation and automated non-destructive harvesting technologies in safflower operations.

Abstract Forecasting stock movements is a crucial research endeavor in finance, aiding traders in making informed decisions for enhanced profitability. Utilizing actual stock prices and correlating factors from the Wind platform presents a potent yet intricate forecasting approach. While previous methodologies have explored this avenue, they encounter challenges including limited comprehension of interrelations among stock data elements, diminished accuracy in extensive series, and struggles with anomaly points. This paper introduces an advanced hybrid model for stock price prediction, termed PMANet. PMANet is founded on Multi-scale Timing Feature Attention, amalgamating Multi-scale Timing Feature Convolution and Ant Particle Swarm Optimization. The model elevates the understanding of dependencies and interrelations within stock data sequences through Probabilistic Positional Attention. Furthermore, the Encoder incorporates Multi-scale Timing Feature Convolution, augmenting the model's capacity to discern multi-scale and significant features while adeptly managing lengthy input sequences. Additionally, the model's proficiency in addressing anomaly points in stock sequences is enhanced by substituting the optimizer with Ant Particle Swarm Optimization. To ascertain the model’s efficacy and applicability, we conducted an empirical study using stocks from four pivotal industries in China. The experimental outcomes demonstrate that PMANet is both feasible and versatile in its predictive capability, yielding forecasts closely aligned with actual values, thereby fulfilling application requirements more effectively.

Abstract Chemical molecular structures are a direct and convenient means of expressing chemical knowledge, playing a vital role in academic communication. In chemistry, hand drawing is a common task for students and researchers. If we can convert hand-drawn chemical molecular structures into machine-readable formats, like SMILES encoding, computers can efficiently process and analyze these structures, significantly enhancing the efficiency of chemical research. Furthermore, with the progress of educational technology, automated grading is gaining popularity. When machines automatically recognize chemical molecular structures and assess the correctness of the drawings, it offers great convenience to teachers. We created ChemReco, a tool designed to identify chemical molecular structures involving three atoms: C, H, and O, providing convenience for chemical researchers. Currently, there are limited studies on hand-drawn chemical molecular structures. Therefore, the primary focus of this paper is constructing datasets. We propose a synthetic image method to rapidly generate images resembling hand-drawn chemical molecular structures, enhancing dataset acquisition efficiency. Regarding model selection, the hand-drawn chemical molecule structural recognition model developed in this article achieves a final recognition accuracy of 96.90%. This model employs the encoder-decoder architecture of EfficientNet + Transformer, demonstrating superior performance compared to other encoder-decoder combinations.

Abstract Handwritten Text Recognition (HTR) is a challenging task due to the complex structures and variations present in handwritten text. In recent years, the application of gated mechanisms, such as Long Short-Term Memory (LSTM) networks, has brought significant advancements to HTR systems. This paper presents an overview of HTR using a gated mechanism and highlights its novelty and advantages. The gated mechanism enables the model to capture long-term dependencies, retain relevant context, handle variable length sequences, mitigate error propagation, and adapt to contextual variations. The pipeline involves preprocessing the handwritten text images, extracting features, modeling the sequential dependencies using the gated mechanism, and decoding the output into readable text. The training process utilizes annotated datasets and optimization techniques to minimize transcription discrepancies. HTR using a gated mechanism has found applications in digitizing historical documents, automatic form processing, and real-time transcription. The results show improved accuracy and robustness compared to traditional HTR approaches. The advancements in HTR using a gated mechanism open up new possibilities for effectively recognizing and transcribing handwritten text in various domains. This research does a better job than the most recent iteration of the HTR system when compared to five different handwritten datasets (Washington, Saint Gall, RIMES, Bentham and IAM). Smartphones and robots are examples of low-cost computing devices that can benefit from this research.

Abstract With technological innovations, enterprises in the real world are managing every iota of data as it can be mined to derive business intelligence (BI). However, when data comes from multiple sources, it may result in duplicate records. As data is given paramount importance, it is also significant to eliminate duplicate entities towards data integration, performance and resource optimization. To realize reliable systems for record deduplication, late, deep learning could offer exciting provisions with a learning-based approach. Deep ER is one of the deep learning-based methods used recently for dealing with the elimination of duplicates in structured data. Using it as a reference model, in this paper, we propose a framework known as Enhanced Deep Learning-based Record Deduplication (EDL-RD) for improving performance further. Towards this end, we exploited a variant of Long Short Term Memory (LSTM) along with various attribute compositions, similarity metrics, and numerical and null value resolution. We proposed an algorithm known as Efficient Learning based Record Deduplication (ELbRD). The algorithm extends the reference model with the aforementioned enhancements. An empirical study has revealed that the proposed framework with extensions outperforms existing methods.

Abstract Due to the large computational overhead, underutilization of features, and high bandwidth consumption in traditional SDN environments for DDoS attack detection and mitigation methods, this paper proposes a two-stage detection and mitigation method for DDoS attacks in SDN based on multi-dimensional characteristics. Firstly, an analysis of the traffic statistics from the SDN switch ports is performed, which aids in conducting a coarse-grained detection of DDoS attacks within the network. Subsequently, a Multi-Dimensional Deep Convolutional Classifier (MDDCC) is constructed using wavelet decomposition and convolutional neural networks to extract multi-dimensional characteristics from the traffic data passing through suspicious switches. Based on these extracted multi-dimensional characteristics, a simple classifier can be employed to accurately detect attack samples. Finally, by integrating graph theory with restrictive strategies, the source of attacks in SDN networks can be effectively traced and isolated. The experimental results indicate that the proposed method, which utilizes a minimal amount of statistical information, can quickly and accurately detect attacks within the SDN network. It demonstrates superior accuracy and generalization capabilities compared to traditional detection methods, especially when tested on both simulated and public datasets. Furthermore, by isolating the affected nodes, the method effectively mitigates the impact of the attacks, ensuring the normal transmission of legitimate traffic during network attacks. This approach not only enhances the detection capabilities but also provides a robust mechanism for containing the spread of cyber threats, thereby safeguarding the integrity and performance of the network.

Abstract With the advancement of science and technology, coal-washing plants are transitioning to intelligent, information-based, and professional sorting systems. This shift accelerates the construction a modern economic system characterized by green and low-carbon development, thereby promoting the high-quality advancement of the coal industry. Traditional manual gangue picking and multi-axis robotic arm gangue selection currently suffer from low recognition accuracy, slow sorting efficiency, and high worker labor intensity. This paper proposes a deep learning-based, non-contact gangue recognition and pneumatic intelligent sorting system. The system constructs a dynamic database containing key feature information such as the target gangue's contour, quality, and center of mass. The system elucidates the relationships between ejection speed, mass, volume, angle of incidence, and the impact energy matching mechanism. Demonstration experiments using the system prototype for coal gangue sorting reveal that, compared to existing robotic arm sorting methods in coal washing plants, this system achieves a gangue identification accuracy exceeding 97%, a sorting rate above 91%, and a separation time of less than 3 s from identification to separation, thereby effectively enhancing raw coal purity.

Abstract The educational environment plays a vital role in the development of students who participate in athletic pursuits both in terms of their physical health and their ability to detect fatigue. As a result of recent advancements in deep learning and biosensors benefitting from edge computing resources, we are now able to monitor the physiological fatigue of students participating in sports in real time. These devices can then be used to analyze the data using contemporary technology. In this paper, we present an innovative deep learning framework for forecasting fatigue in athletic students following physical exercise. It addresses the issue of lack of precision computational models and extensive data analysis in current approaches to monitoring students’ physical activity. In our study, we classified fatigue and non-fatigue based on photoplethysmography (PPG) signals. Several deep learning models are compared in the study. Using limited training data, determining the optimal parameters for PPG presents a significant challenge. For datasets containing many data points, several models were trained using PPG signals: a deep residual network convolutional neural network (ResNetCNN) ResNetCNN, an Xception architecture, a bidirectional long short-term memory (BILSTM), and a combination of these models. Training and testing datasets were assigned using a fivefold cross validation approach. Based on the testing dataset, the model demonstrated a proper classification accuracy of 91.8%.

Abstract Alzheimer’s disease (AD), the predominant form of dementia, is a growing global challenge, emphasizing the urgent need for accurate and early diagnosis. Current clinical diagnoses rely on radiologist expert interpretation, which is prone to human error. Deep learning has thus far shown promise for early AD diagnosis. However, existing methods often overlook focal structural atrophy critical for enhanced understanding of the cerebral cortex neurodegeneration. This paper proposes a deep learning framework that includes a novel structure-focused neurodegeneration CNN architecture named SNeurodCNN and an image brightness enhancement preprocessor using gamma correction. The SNeurodCNN architecture takes as input the focal structural atrophy features resulting from segmentation of brain structures captured through magnetic resonance imaging (MRI). As a result, the architecture considers only necessary CNN components, which comprises of two downsampling convolutional blocks and two fully connected layers, for achieving the desired classification task, and utilises regularisation techniques to regularise learnable parameters. Leveraging mid-sagittal and para-sagittal brain image viewpoints from the Alzheimer’s disease neuroimaging initiative (ADNI) dataset, our framework demonstrated exceptional performance. The para-sagittal viewpoint achieved 97.8% accuracy, 97.0% specificity, and 98.5% sensitivity, while the mid-sagittal viewpoint offered deeper insights with 98.1% accuracy, 97.2% specificity, and 99.0% sensitivity. Model analysis revealed the ability of SNeurodCNN to capture the structural dynamics of mild cognitive impairment (MCI) and AD in the frontal lobe, occipital lobe, cerebellum, temporal, and parietal lobe, suggesting its potential as a brain structural change digi-biomarker for early AD diagnosis. This work can be reproduced using code we made available on GitHub.

Abstract The hippocampus is a critical component of the brain and is associated with many neurological disorders. It can be further subdivided into several subfields, and accurate segmentation of these subfields is of great significance for diagnosis and research. However, the structures of hippocampal subfields are irregular and have complex boundaries, and their voxel values are close to surrounding brain tissues, making the segmentation task highly challenging. Currently, many automatic segmentation tools exist for hippocampal subfield segmentation, but they suffer from high time costs and low segmentation accuracy. In this paper, we propose a new dual-branch segmentation network structure (DSnet) based on deep learning for hippocampal subfield segmentation. While traditional convolutional neural network-based methods are effective in capturing hierarchical structures, they struggle to establish long-term dependencies. The DSnet integrates the Transformer architecture and a hybrid attention mechanism, enhancing the network’s global perceptual capabilities. Moreover, the dual-branch structure of DSnet leverages the segmentation results of the hippocampal region to facilitate the segmentation of its subfields. We validate the efficacy of our algorithm on the public Kulaga-Yoskovitz dataset. Experimental results indicate that our method is more effective in segmenting hippocampal subfields than conventional single-branch network structures. Compared to the classic 3D U-Net, our proposed DSnet improves the average Dice accuracy of hippocampal subfield segmentation by 0.57%.

Abstract Early detection of the adenocarcinoma cancer in colon tissue by means of explainable deep learning, by classifying histological images and providing visual explainability on model prediction. Considering that in recent years, deep learning techniques have emerged as powerful techniques in medical image analysis, offering unprecedented accuracy and efficiency, in this paper we propose a method to automatically detect the presence of cancerous cells in colon tissue images. Various deep learning architectures are considered, with the aim of considering the best one in terms of quantitative and qualitative results. As a matter of fact, we consider qualitative results by taking into account the so-called prediction explainability, by providing a way to highlight on the tissue images the areas that from the model point of view are related to the presence of colon cancer. The experimental analysis, performed on 10,000 colon issue images, showed the effectiveness of the proposed method by obtaining an accuracy equal to 0.99. The experimental analysis shows that the proposed method can be successfully exploited for colon cancer detection and localisation from tissue images.

Abstract As the mechanization of the CBM extraction process advances and geological conditions continuously evolve, the production data from CBM wells is deviating increasingly from linearity, thereby presenting a significant challenge in accurately predicting future gas production from these wells. When it comes to predicting the production of CBM, a single deep-learning model can face several drawbacks such as overfitting, gradient explosion, and gradient disappearance. These issues can ultimately result in insufficient prediction accuracy, making it important to carefully consider the limitations of any given model. It’s impressive to see how advanced technology can enhance the prediction accuracy of CBM. In this paper, the use of a CNN model to extract features from CBM well data and combine it with Bi-LSTM and a Multi-Head Attention mechanism to construct a production prediction model for CBM wells—the CNN-BL-MHA model—is fascinating. It is even more exciting that predictions of gas production for experimental wells can be conducted using production data from Wells W1 and W2 as the model’s database. We compared and analyzed the prediction results obtained from the CNN-BL-MHA model we constructed with those from single models like ARIMA, LSTM, MLP, and GRU. The results show that the CNN-BL-MHA model proposed in the study has shown promising results in improving the accuracy of gas production prediction for CBM wells. It’s also impressive that this model demonstrated super stability, which is essential for reliable predictions. Compared to the single deep learning model used in this study, its prediction accuracy can be improved up to 35%, and the prediction results match the actual yield data with lower error.

Abstract To address the urgent need for agricultural intelligence in the face of increasing agricultural output and a shortage of personnel, this paper proposes a high precision object detection network for automated pear picking tasks. The current object detection method using deep learning does not fully consider the redundant background information of the pear detection scene and the mutual occlusion characteristics of multiple pears, so that the detection accuracy is low and cannot meet the needs of complex automated pear picking detection tasks. The proposed, High-level deformation-perception Network with multi-object search NMS(HDMNet), is based on YOLOv8 and utilizes a high-level Semantic focused attention mechanism module to eliminate irrelevant background information and a deformation-perception feature pyramid network to improve accuracy of long-distance and small scale fruit. A multi-object search non-maximum suppression is also proposed to choose the anchor frame in a combined search method suitable for multiple pears. The experimental results show that the HDMNet parameter amount is as low as 12.9 M, the GFLOPs is 41.1, the mAP is 75.7%, the mAP50 reaches 93.6%, the mAP75 reaches 70.2%, and the FPS reaches 73.0. Compared with other SOTA object detection methods, it has the transcend of real-time detection, low parameter amount, low calculation amount, high precision, and accurate positioning.

Abstract Profiling social media users is an analytical approach to generate an extensive blueprint of user’s personal characteristics, which can be useful for a diverse range of applications, such as targeted marketing and personalized recommendations. Although social user profiling has gained substantial attention in recent years, effectively constructing a collaborative model that could describe long and short-term profiles is still challenging. In this paper, we will discuss the profiling problem from two perspectives; how to mathematically model and track user’s behavior over short and long periods and how to enhance the classification of user’s activities. Using mathematical equations, our model can define periods in which the user's interests abruptly changed. A dataset consisting of 30,000 tweets was built and manually annotated into 10 topic categories. Bi-LSTM and GRU models are applied to classify the user’s activities representing his interests, which then are utilized to create and model the dynamic profile. In addition, the effect of word embedding techniques and pre-trained classification models on the accuracy of the classification process is explored in this research.

Abstract This article aims to improve the deep-learning-based surface defect recognition. In actual manufacturing processes, there are issues such as data imbalance, insufficient diversity, and poor quality of augmented data in the collected image data for product defect recognition. A novel defect generation method with multiple loss functions, DG2GAN is presented in this paper. This method employs cycle consistency loss to generate defect images from a large number of defect-free images, overcoming the issue of imbalanced original training data. DJS optimized discriminator loss is introduced in the added discriminator to encourage the generation of diverse defect images. Furthermore, to maintain diversity in generated images while improving image quality, a new DG2 adversarial loss is proposed with the aim of generating high-quality and diverse images. The experiments demonstrated that DG2GAN produces defect images of higher quality and greater diversity compared with other advanced generation methods. Using the DG2GAN method to augment defect data in the CrackForest and MVTec datasets, the defect recognition accuracy increased from 86.9 to 94.6%, and the precision improved from 59.8 to 80.2%. The experimental results show that using the proposed defect generation method can obtain sample images with high quality and diversity and employ this method for data augmentation significantly enhances surface defect recognition technology.

Abstract Vehicle identification systems are vital components that enable many aspects of contemporary life, such as safety, trade, transit, and law enforcement. They improve community and individual well-being by increasing vehicle management, security, and transparency. These tasks entail locating and extracting license plates from images or video frames using computer vision and machine learning techniques, followed by recognizing the letters or digits on the plates. This paper proposes a new license plate detection and recognition method based on the deep learning YOLO v8 method, image processing techniques, and the OCR technique for text recognition. For this, the first step was the dataset creation, when gathering 270 images from the internet. Afterward, CVAT (Computer Vision Annotation Tool) was used to annotate the dataset, which is an open-source software platform made to make computer vision tasks easier to annotate and label images and videos. Subsequently, the newly released Yolo version, the Yolo v8, has been employed to detect the number plate area in the input image. Subsequently, after extracting the plate the k-means clustering algorithm, the thresholding techniques, and the opening morphological operation were used to enhance the image and make the characters in the license plate clearer before using OCR. The next step in this process is using the OCR technique to extract the characters. Eventually, a text file containing only the character reflecting the vehicle's country is generated. To ameliorate the efficiency of the proposed approach, several metrics were employed, namely precision, recall, F1-Score, and CLA. In addition, a comparison of the proposed method with existing techniques in the literature has been given. The suggested method obtained convincing results in both detection as well as recognition by obtaining an accuracy of 99% in detection and 98% in character recognition.

Abstract Accurate power load forecasting is crucial for the sustainable operation of smart grids. However, the complexity and uncertainty of load, along with the large-scale and high-dimensional energy information, present challenges in handling intricate dynamic features and long-term dependencies. This paper proposes a computational approach to address these challenges in short-term power load forecasting and energy information management, with the goal of accurately predicting future load demand. The study introduces a hybrid method that combines multiple deep learning models, the Gated Recurrent Unit (GRU) is employed to capture long-term dependencies in time series data, while the Temporal Convolutional Network (TCN) efficiently learns patterns and features in load data. Additionally, the attention mechanism is incorporated to automatically focus on the input components most relevant to the load prediction task, further enhancing model performance. According to the experimental evaluation conducted on four public datasets, including GEFCom2014, the proposed algorithm outperforms the baseline models on various metrics such as prediction accuracy, efficiency, and stability. Notably, on the GEFCom2014 dataset, FLOP is reduced by over 48.8%, inference time is shortened by more than 46.7%, and MAPE is improved by 39%. The proposed method significantly enhances the reliability, stability, and cost-effectiveness of smart grids, which facilitates risk assessment optimization and operational planning under the context of information management for smart grid systems.

Abstract The precise identification of surface imperfections in steel strips is crucial for ensuring steel product quality. To address the challenges posed by the substantial model size and computational complexity in current algorithms for detecting surface defects in steel strips, this paper introduces SS-YOLO (YOLOv7 for Steel Strip), an enhanced lightweight YOLOv7 model. This method replaces the CBS module in the backbone network with a lightweight MobileNetv3 network, reducing the model size and accelerating the inference time. The D-SimSPPF module, which integrates depth separable convolution and a parameter-free attention mechanism, was specifically designed to replace the original SPPCSPC module within the YOLOv7 network, expanding the receptive field and reducing the number of network parameters. The parameter-free attention mechanism SimAM is incorporated into both the neck network and the prediction output section, enhancing the ability of the model to extract essential features of strip surface defects and improving detection accuracy. The experimental results on the NEU-DET dataset show that SS-YOLO achieves a 97% mAP50 accuracy, which is a 4.5% improvement over that of YOLOv7. Additionally, there was a 79.3% reduction in FLOPs(G) and a 20.7% decrease in params. Thus, SS-YOLO demonstrates an effective balance between detection accuracy and speed while maintaining a lightweight profile.

Abstract In the dynamic urban landscape, understanding the distribution of buildings is paramount. Extracting and delineating building footprints from high-resolution images, captured by aerial platforms or satellites, is essential but challenging to accomplish manually, due to the abundance of high-resolution data. Automation becomes imperative, yet it introduces complexities related to handling diverse data sources and the computational demands of advanced algorithms. The innovative solution proposed in this paper addresses some intricate challenges occurring when integrating deep learning and data fusion on Earth Observed imagery. By merging RGB orthophotos with Digital Surface Models, deriving from the same aerial high-resolution surveys, an integrated consistent four-band dataset is generated. This unified approach, focused on the extraction of height information through stereoscopy utilizing a singular source, facilitates enhanced pixel-to-pixel data fusion. Employing DeepLabv3 algorithms, a state-of-the-art semantic segmentation network for multi-scale context, pixel-based segmentation on the integrated dataset was performed, excelling in capturing intricate details, particularly when enhanced by the additional height information deriving from the Digital Surface Models acquired over urban landscapes. Evaluation over a 21 km2 area in Turin, Italy, featuring diverse building frameworks, showcases how the proposed approach leads towards superior accuracy levels and building boundary refinement. Notably, the methodology discussed in the present article, significantly reduces training time compared to conventional approaches like U-Net, overcoming inherent challenges in high-resolution data automation. By establishing the effectiveness of leveraging DeepLabv3 algorithms on an integrated dataset for precise building footprint segmentation, the present contribution holds promise for applications in 3D modelling, Change detection and urban planning. An approach favouring the application of deep learning strategies on integrated high-resolution datasets can then guide decision-making processes facilitating urban management tasks.

Abstract Hardware Trojans (HTs) are hidden threats embedded in the circuitry of integrated circuits (ICs), enabling unauthorized access, data theft, operational disruptions, or even physical harm. Detecting Hardware Trojans (HTD) is paramount for ensuring IC security. This paper introduces a novel Siamese neural network (SNN) framework for non-destructive HTD. The proposed framework can detect HTs by processing power side-channel signals without the need for a golden model of the IC. To obtain the best results, different neural network models such as Convolutional Neural Network (CNN), Gated Recurrent Unit (GRU), and Long Short-Term Memory (LSTM) are integrated individually with SNN. These models are trained on the extracted features from the Trojan Power & EM Side-Channel dataset. The results show that the Siamese LSTM model achieved the highest accuracy of 86.78%, followed by the Siamese GRU model with 83.59% accuracy and the Siamese CNN model with 73.54% accuracy. The comparison shows that of the proposed Siamese LSTM is a promising new approach for HTD and outperform the state-of-the-art methods.

Abstract In recent years, the growth spurt of medical imaging data has led to the development of various machine learning algorithms for various healthcare applications. The MedMNISTv2 dataset, a comprehensive benchmark for 2D biomedical image classification, encompasses diverse medical imaging modalities such as Fundus Camera, Breast Ultrasound, Colon Pathology, Blood Cell Microscope etc. Highly accurate classifications performed on these datasets is crucial for identification of various diseases and determining the course of treatment. This research paper presents a comprehensive analysis of four subsets within the MedMNISTv2 dataset: BloodMNIST, BreastMNIST, PathMNIST and RetinaMNIST. Each of these selected datasets is of diverse data modalities and comes with various sample sizes, and have been selected to analyze the efficiency of the model against diverse data modalities. The study explores the idea of assessing the Vision Transformer Model’s ability to capture intricate patterns and features crucial for these medical image classification and thereby transcend the benchmark metrics substantially. The methodology includes pre-processing the input images which is followed by training the ViT-base-patch16-224 model on the mentioned datasets. The performance of the model is assessed using key metrices and by comparing the classification accuracies achieved with the benchmark accuracies. With the assistance of ViT, the new benchmarks achieved for BloodMNIST, BreastMNIST, PathMNIST and RetinaMNIST are 97.90%, 90.38%, 94.62% and 57%, respectively. The study highlights the promise of Vision transformer models in medical image analysis, preparing the way for their adoption and further exploration in healthcare applications, aiming to enhance diagnostic accuracy and assist medical professionals in clinical decision-making.

Abstract The resolution of traffic congestion and personal safety issues holds paramount importance for human’s life. The ability of an autonomous driving system to navigate complex road conditions is crucial. Deep learning has greatly facilitated machine vision perception in autonomous driving. Aiming at the problem of small target detection in traditional YOLOv5s, this paper proposes an optimized target detection algorithm. The C3 module on the algorithm’s backbone is upgraded to the CBAMC3 module, introducing a novel GELU activation function and EfficiCIoU loss function, which accelerate convergence on position loss l box , confidence loss l obj , and classification loss l cls , enhance image learning capabilities and address the issue of inaccurate detection of small targets by improving the algorithm. Testing with a vehicle-mounted camera on a predefined route effectively identifies road vehicles and analyzes depth position information. The avoidance model, combined with Pure Pursuit and MPC control algorithms, exhibits more stable variations in vehicle speed, front-wheel steering angle, lateral acceleration, etc., compared to the non-optimized version. The robustness of the driving system's visual avoidance functionality is enhanced, further ameliorating congestion issues and ensuring personal safety.

Abstract Recent research has focused extensively on employing Deep Learning (DL) techniques, particularly Convolutional Neural Networks (CNN), for Speech Emotion Recognition (SER). This study addresses the burgeoning interest in leveraging DL for SER, specifically focusing on Punjabi language speakers. The paper presents a novel approach to constructing and preprocessing a labeled speech corpus using diverse social media sources. By utilizing spectrograms as the primary feature representation, the proposed algorithm effectively learns discriminative patterns for emotion recognition. The method is evaluated on a custom dataset derived from various Punjabi media sources, including films and web series. Results demonstrate that the proposed approach achieves an accuracy of 69%, surpassing traditional methods like decision trees, Naïve Bayes, and random forests, which achieved accuracies of 49%, 52%, and 61% respectively. Thus, the proposed method improves accuracy in recognizing emotions from Punjabi speech signals.

Abstract The railway rivet is one of the most important and easily damaged parts of the connection. If rivets develop cracks during the production process, their load-bearing capacity will be reduced, thereby increasing the risk of failure. Fluorescent magnetic particle flaw detection (FMPFD) is a widely used inspection method for train fasteners. Manual inspection is not only time-consuming but also prone to miss detection, therefore intelligent detection system has important application value. However, the fluorescent crack images obtained by FMPFD present challenges for intelligent detection, such as the dense, multi-scaled and uninstantiated cracks. In addition, there is limited research on fluorescent rivet crack detection. This paper adopts instance segmentation to achieve automatic cracks detection of rivets. A decentralized target center and low overlap rate labeling method is proposed, and a Gaussian-weighted correction post-processing method is introduced to improve the recall rate in the areas of dense cracks. An efficient channel spatial attention mechanism for feature extraction is proposed in order to enhance the detection of multi-scale cracks. For uninstantiated cracks, an improvement of crack detection in uninstantiated regions based on multi task feature learning is proposed, thoroughly utilizing the semantic and spatial features of the fluorescent cracks. The experimental results show that the improved methods are better than the baseline and some cutting-edge algorithms, achieving a recall rate and mAP0.5 of 86.4% and 90.3%. In addition, a single coil non-contact train rivet composite magnetization device is built for rivets that can magnetize different shapes of rivets and has universality.

Abstract This paper aims to propose a prediction method based on Deep Learning (DL) and Internet of Things (IoT) technology, focusing on the ecological security and tourist satisfaction of Ice-and-Snow Tourism (IST) to solve practical problems in this field. Accurate predictions of ecological security and tourist satisfaction in IST have been achieved by collecting and analyzing environment and tourist behavior data and combining with DL models, such as convolutional and recurrent neural networks. The experimental results show that the proposed method has significant advantages in performance indicators, such as accuracy, F1 score, Mean Squared Error (MSE), and correlation coefficient. Compared to other similar methods, the method proposed improves accuracy by 3.2%, F1 score by 0.03, MSE by 0.006, and correlation coefficient by 0.06. These results emphasize the important role of combining DL with IoT technology in predicting ecological security and tourist satisfaction in IST.

Abstract The challenge of developing an Android malware detection framework that can identify malware in real-world apps is difficult for academicians and researchers. The vulnerability lies in the permission model of Android. Therefore, it has attracted the attention of various researchers to develop an Android malware detection model using permission or a set of permissions. Academicians and researchers have used all extracted features in previous studies, resulting in overburdening while creating malware detection models. But, the effectiveness of the machine learning model depends on the relevant features, which help in reducing the value of misclassification errors and have excellent discriminative power. A feature selection framework is proposed in this research paper that helps in selecting the relevant features. In the first stage of the proposed framework, t-test, and univariate logistic regression are implemented on our collected feature data set to classify their capacity for detecting malware. Multivariate linear regression stepwise forward selection and correlation analysis are implemented in the second stage to evaluate the correctness of the features selected in the first stage. Furthermore, the resulting features are used as input in the development of malware detection models using three ensemble methods and a neural network with six different machine-learning algorithms. The developed models’ performance is compared using two performance parameters: F-measure and Accuracy. The experiment is performed by using half a million different Android apps. The empirical findings reveal that malware detection model developed using features selected by implementing proposed feature selection framework achieved higher detection rate as compared to the model developed using all extracted features data set. Further, when compared to previously developed frameworks or methodologies, the experimental results indicates that model developed in this study achieved an accuracy of 98.8%.

Abstract Agricultural dykelands in Nova Scotia rely heavily on a surface drainage technique called land forming, which is used to alter the topography of fields to improve drainage. The presence of land-formed fields provides useful information to better understand land utilization on these lands vulnerable to rising sea levels. Current field boundaries delineation and classification methods, such as manual digitalization and traditional segmentation techniques, are labour-intensive and often require manual and time-consuming parameter selection. In recent years, deep learning (DL) techniques, including convolutional neural networks and Mask R-CNN, have shown promising results in object recognition, image classification, and segmentation tasks. However, there is a gap in applying these techniques to detecting surface drainage patterns on agricultural fields. This paper develops and tests a Mask R-CNN model for detecting land-formed fields on agricultural dykelands using LiDAR-derived elevation data. Specifically, our approach focuses on identifying groups of pixels as cohesive objects within the imagery, a method that represents a significant advancement over pixel-by-pixel classification techniques. The DL model developed in this study demonstrated a strong overall performance, with a mean Average Precision (mAP) of 0.89 across Intersection over Union (IoU) thresholds from 0.5 to 0.95, indicating its effectiveness in detecting land-formed fields. Results also revealed that 53% of Nova Scotia’s dykelands are being used for agricultural purposes and approximately 75% (6924 hectares) of these fields were land-formed. By applying deep learning techniques to LiDAR-derived elevation data, this study offers novel insights into surface drainage mapping, enhancing the capability for precise and efficient agricultural land management in regions vulnerable to environmental changes.

Abstract Fairness has become a critical value online, and the latest studies consider it in many problems. In recommender systems, fairness is important since the visibility of items is controlled by systems. Previous fairness-aware recommender systems assume that sufficient relationship data between users and items are available. However, it is common that new users and items are frequently introduced, and they have no relationship data yet. In this paper, we study recommendation methods to enhance fairness in a cold-start state. Fairness is more significant when the preference of a user or the popularity of an item is unknown. We propose a meta-learning-based cold-start recommendation framework called FaRM to alleviate the unfairness of recommendations. The proposed framework consists of three steps. We first propose a fairness-aware meta-path generation method to eliminate bias in sensitive attributes. In addition, we construct fairness-aware user representations through the meta-path aggregation approach. Then, we propose a novel fairness objective function and introduce a joint learning method to minimize the trade-off between relevancy and fairness. In extensive experiments with various cold-start scenarios, it is shown that FaRM is significantly superior in fairness performance while preserving relevance accuracy over previous work.

Abstract The objective of human pose estimation (HPE) derived from deep learning aims to accurately estimate and predict the human body posture in images or videos via the utilization of deep neural networks. However, the accuracy of real-time HPE tasks is still to be improved due to factors such as partial occlusion of body parts and limited receptive field of the model. To alleviate the accuracy loss caused by these issues, this paper proposes a real-time HPE model called $${\textbf {CCAM-Person}}$$ CCAM - Person based on the YOLOv8 framework. Specifically, we have improved the backbone and neck of the YOLOv8x-pose real-time HPE model to alleviate the feature loss and receptive field constraints. Secondly, we introduce the context coordinate attention module (CCAM) to augment the model’s focus on salient features, reduce background noise interference, alleviate key point regression failure caused by limb occlusion, and improve the accuracy of pose estimation. Our approach attains competitive results on multiple metrics of two open-source datasets, MS COCO 2017 and CrowdPose. Compared with the baseline model YOLOv8x-pose, CCAM-Person improves the average precision by 2.8% and 3.5% on the two datasets, respectively.

Abstract Many intrusion detection techniques have been developed to ensure that the target system can function properly under the established rules. With the booming Internet of Things (IoT) applications, the resource-constrained nature of its devices makes it urgent to explore lightweight and high-performance intrusion detection models. Recent years have seen a particularly active application of deep learning (DL) techniques. The spiking neural network (SNN), a type of artificial intelligence that is associated with sparse computations and inherent temporal dynamics, has been viewed as a potential candidate for the next generation of DL. It should be noted, however, that current research into SNNs has largely focused on scenarios where limited computational resources and insufficient power sources are not considered. Consequently, even state-of-the-art SNN solutions tend to be inefficient. In this paper, a lightweight and effective detection model is proposed. With the help of rational algorithm design, the model integrates the advantages of SNNs as well as convolutional neural networks (CNNs). In addition to reducing resource usage, it maintains a high level of classification accuracy. The proposed model was evaluated against some current state-of-the-art models using a comprehensive set of metrics. Based on the experimental results, the model demonstrated improved adaptability to environments with limited computational resources and energy sources.

Abstract Sensors, wearables, mobile devices, and other Internet of Things (IoT) devices are becoming increasingly integrated into all aspects of our lives. They are capable of gathering enormous amounts of data, such as image data, which can then be sent to the cloud for processing. However, this results in an increase in network traffic and latency. To overcome these difficulties, edge computing has been proposed as a paradigm for computing that brings processing closer to the location where data is produced. This paper explores the merging of cloud and edge computing for IoT and investigates approaches using machine learning for dimensionality reduction of images on the edge, employing the autoencoder deep learning-based approach and principal component analysis (PCA). The encoded data is then sent to the cloud server, where it is used directly for any machine learning task without significantly impacting the accuracy of the data processed in the cloud. The proposed approach has been evaluated on an object detection task using a set of 4000 images randomly chosen from three datasets: COCO, human detection, and HDA datasets. Results show that a 77% reduction in data did not have a significant impact on the object detection task’s accuracy.

Abstract Identifying and recognizing the food on the basis of its eating sounds is a challenging task, as it plays an important role in avoiding allergic foods, providing dietary preferences to people who are restricted to a particular diet, showcasing its cultural significance, etc. In this research paper, the aim is to design a novel methodology that helps to identify food items by analyzing their eating sounds using various deep learning models. To achieve this objective, a system has been proposed that extracts meaningful features from food-eating sounds with the help of signal processing techniques and deep learning models for classifying them into their respective food classes. Initially, 1200 audio files for 20 food items labeled have been collected and visualized to find relationships between the sound files of different food items. Later, to extract meaningful features, various techniques such as spectrograms, spectral rolloff, spectral bandwidth, and mel-frequency cepstral coefficients are used for the cleaning of audio files as well as to capture the unique characteristics of different food items. In the next phase, various deep learning models like GRU, LSTM, InceptionResNetV2, and the customized CNN model have been trained to learn spectral and temporal patterns in audio signals. Besides this, the models have also been hybridized i.e. Bidirectional LSTM + GRU and RNN + Bidirectional LSTM, and RNN + Bidirectional GRU to analyze their performance for the same labeled data in order to associate particular patterns of sound with their corresponding class of food item. During evaluation, the highest accuracy, precision,F1 score, and recall have been obtained by GRU with 99.28%, Bidirectional LSTM + GRU with 97.7% as well as 97.3%, and RNN + Bidirectional LSTM with 97.45%, respectively. The results of this study demonstrate that deep learning models have the potential to precisely identify foods on the basis of their sound by computing the best outcomes.

Abstract Internet of Things (IoT) integration in healthcare improves patient care while also making healthcare delivery systems more effective and economical. To fully realize the advantages of IoT in healthcare, it is imperative to overcome issues with data security, interoperability, and ethical considerations. IoT sensors periodically measure the health-related data of the patients and share it with a server for further evaluation. At the server, different machine learning algorithms are applied which help in early diagnosis of diseases and issue alerts in case vital signs are out of the normal range. Different cyber attacks can be launched on IoT devices which can result in compromised security and privacy of applications such as health care. In this paper, we utilize the publicly available Canadian Institute for Cybersecurity (CIC) IoT dataset to model machine learning techniques for efficient detection of anomalous network traffic. The dataset consists of 33 types of IoT attacks which are divided into 7 main categories. In the current study, the dataset is pre-processed, and a balanced representation of classes is used in generating a non-biased supervised (Random Forest, Adaptive Boosting, Logistic Regression, Perceptron, Deep Neural Network) machine learning models. These models are analyzed further by eliminating highly correlated features, reducing dimensionality, minimizing overfitting, and speeding up training times. Random Forest was found to perform optimally across binary and multiclass classification of IoT Attacks with an approximate accuracy of 99.55% under both reduced and all feature space. This improvement was complimented by a reduction in computational response time which is essential for real-time attack detection and response.

Abstract Parasitic organisms pose a major global health threat, mainly in regions that lack advanced medical facilities. Early and accurate detection of parasitic organisms is vital to saving lives. Deep learning models have uplifted the medical sector by providing promising results in diagnosing, detecting, and classifying diseases. This paper explores the role of deep learning techniques in detecting and classifying various parasitic organisms. The research works on a dataset consisting of 34,298 samples of parasites such as Toxoplasma Gondii, Trypanosome, Plasmodium, Leishmania, Babesia, and Trichomonad along with host cells like red blood cells and white blood cells. These images are initially converted from RGB to grayscale followed by the computation of morphological features such as perimeter, height, area, and width. Later, Otsu thresholding and watershed techniques are applied to differentiate foreground from background and create markers on the images for the identification of regions of interest. Deep transfer learning models such as VGG19, InceptionV3, ResNet50V2, ResNet152V2, EfficientNetB3, EfficientNetB0, MobileNetV2, Xception, DenseNet169, and a hybrid model, InceptionResNetV2, are employed. The parameters of these models are fine-tuned using three optimizers: SGD, RMSprop, and Adam. Experimental results reveal that when RMSprop is applied, VGG19, InceptionV3, and EfficientNetB0 achieve the highest accuracy of 99.1% with a loss of 0.09. Similarly, using the SGD optimizer, InceptionV3 performs exceptionally well, achieving the highest accuracy of 99.91% with a loss of 0.98. Finally, applying the Adam optimizer, InceptionResNetV2 excels, achieving the highest accuracy of 99.96% with a loss of 0.13, outperforming other optimizers. The findings of this research signify that using deep learning models coupled with image processing methods generates a highly accurate and efficient way to detect and classify parasitic organisms.

Abstract All-optical plasmonic switches (AOPSs) utilizing surface plasmon polaritons are well-suited for integration into photonic integrated circuits (PICs) and play a crucial role in advancing all-optical signal processing. The current AOPS design methods still rely on trial-and-error or empirical approaches. In contrast, recent deep learning (DL) advances have proven highly effective as computational tools, offering an alternative means to accelerate nanophotonics simulations. This paper proposes an innovative approach utilizing DL for spectrum prediction and inverse design of AOPS. The switches employ circular nonlinear plasmonic ring resonators (NPRRs) composed of interconnected metal–insulator–metal waveguides with a ring resonator. The NPRR switching performance is shown using the nonlinear Kerr effect. The forward model presented in this study demonstrates superior computational efficiency when compared to the finite-difference time-domain method. The model analyzes various structural parameters to predict transmission spectra with a distinctive dip. Inverse modeling enables the prediction of design parameters for desired transmission spectra. This model provides a rapid estimation of design parameters, offering a clear advantage over time-intensive conventional optimization approaches. The loss of prediction for both the forward and inverse models, when compared to simulations, is exceedingly low and on the order of 10−4. The results confirm the suitability of employing DL for forward and inverse design of AOPSs in PICs.