Numerous image processing techniques (IPTs) have been employed to detect crack defects, offering an alternative to human-conducted onsite inspections. These IPTs manipulate images to extract defect features, particularly cracks in surfaces produced through Additive Manufacturing (AM). This article presents a vision-based approach that utilizes deep convolutional neural networks (CNNs) for crack detection in AM surfaces. Traditional image processing techniques face challenges with diverse real-world scenarios and varying crack types. To overcome these challenges, our proposed method leverages CNNs, eliminating the need for extensive feature extraction. Annotation for CNN training is facilitated by LabelImg without the requirement for additional IPTs. The trained CNN, enhanced by OpenCV preprocessing techniques, achieves an outstanding 99.54% accuracy on a dataset of 14,982 annotated images with resolutions of 1536 × 1103 pixels. Evaluation metrics exceeding 96% precision, 98% recall, and a 97% F1-score highlight the precision and effectiveness of the entire process.

The information of seismic damage of buildings in SAR images of different time phase, especially in SAR images after earthquake, is easily disturbed by other factors, which affects the accuracy of information discrimination. In order to identify and evaluate the distribution information of the seismic damage accurately and make full use of the abundant texture features in the SAR image. The conventional method of change detection based on texture features usually takes the pixel as the calculating unit. In this paper, a method of texture feature change detection of SAR images based on watershed segmentation algorithm is proposed. Based on the optimization of texture feature parameters, the feature parameters are segmented by the watershed segmentation algorithm, and the feature object image is obtained. This method introduces the idea of object oriented, and carries out the calculation of the difference map at the object level, Finally, the classification threshold value of different types of seismic damage types is selected, and the recognition of building damage is achieved. Taking the ALOS data before and after the earthquake in Yushu as an example to verify the effectiveness of the method, the overall accuracy of the building extraction is 88.9%, Compared with pixel-based methods, it is proved that the proposed method is effective.

Detecting cracks in structural health monitoring is crucial for ensuring infrastructure safety and longevity. Using drones to obtain crack images and automate processing can improve the efficiency of crack detection. To address the challenges posed by the limited computing resources of edge devices in practical applications, we propose CrackScopeNet, a lightweight segmentation network model that simultaneously considers local and global crack features, suitable for deployment on drone platforms with limited computational power and memory. CrackScopeNet features a multi-scale branch to improve sensitivity to cracks of varying sizes without substantial computational overhead and a stripe-wise context attention mechanism to enhance the capture of long-range contextual information, mitigating the interference from complex backgrounds. Experimental results on CrackSeg9k dataset demonstrate that CrackScopeNet leads to a significant improvement in prediction performance, with the highest mean intersection over union (mIoU) scores reaching 82.12%, while maintaining a lightweight architecture with only 1.05M parameters and 1.58G floating point operations (FLOPs). Besides, CrackScopeNet excels in inference speed on edge devices without GPU because of its low FLOPs. CrackScopeNet contributes to the development of efficient and effective crack segmentation networks, suitable for practical structural health monitoring applications using drone platforms.

Explosion monitoring is performed by infrasound and seismoacoustic sensor networks that are distributed globally, regionally, and locally. However, these networks are unevenly and sparsely distributed, especially in the local scale as maintaining and deploying networks is costly. With increasing interest in smaller yield explosions, the need for more dense networks has increased. To address this issue, we propose using smartphone sensors for explosion detection as they are cost-effective and easy to deploy. Although there are studies using smartphone sensors for explosion detection, the field is still in its infancy and new technologies need to be developed. We applied a machine learning model for explosion detection using smartphone microphones. The data used were from the Smartphone High-explosion Audio Recordings Dataset (SHAReD), a collection of 326 waveforms from 70 high-explosive (HE) events recorded on smartphones, and the ESC-50 dataset, a benchmarking dataset commonly used for environmental sound classification. Two machine learning models were trained and combined into an ensemble model for explosion detection. The resulting ensemble model classified audio signals as either “explosion,” “ambient,” or “other” with true positive rates (recall) greater than 96% for all three categories.

Landslides, resulting from disturbances in slope equilibrium, pose a significant threat to landscapes, infrastructure, and human life. Triggered by factors such as intense precipitation, seismic activities, or volcanic eruptions, these events can cause extensive damage and endanger nearby communities. A comprehensive understanding of landslide characteristics, including spatio-temporal patterns, dimensions, and morphology, is vital for effective landslide-disaster management. Existing Remote Sensing approaches mostly use either optical or Synthetic Aperture Radar (SAR) sensors. Integrating information from both these types of sensors promises greater accuracy for identifying and locating landslides. This study proposes a novel approach, the ML-LaDeCORsat (Machine Learning-based coseismic Landslide Detection using Combined Optical and Radar Satellite Imagery), that integrates freely available Sentinel-1, Palsar-2, and Sentinel-2 imagery data, along with relevant spectral indices and suitable bands using machine-learning-based classification for coseismic landslide detection implemented in Google Earth Engine (GEE). The approach includes a robust and reproducible training and validation strategy. Using landslides from four different earthquake case studies, we demonstrate the superiority of our approach over existing solutions in coseismic landslide iden-tification and localization, providing a detection accuracy of 87-92%. ML-LaDeCORsat can be adapted to other landslide events (GEE script is provided). Transfer learning experiments proofed that our model can be applied to other coseismic landslide events without the need for additional training data. Our novel approach therefore facilitates quick and reliable identification of co-seismic landslides, highlighting its potential to contribute towards more effective disaster management.

Lane detection is a foundational technology for autonomous driving systems. It involves identifying the boundaries of lanes on the road and ensuring the vehicle stays within these boundaries. Accurate lane detection is crucial for safety, navigation and traffic management. This paper presents an artificial intelligent based autonomous road lane detection and navigation system for vehicles. The system can detect and analyze road lanes accurately and efficiently in real-time, with frame rates of up to 17 FPS. By utilizing image processing techniques, the system can identify the location and boundaries of road lanes and obstacles on the road and provide accurate and reliable navigation guidance to the driver. The system is integrated with sensors and actuators to provide comprehensive autonomous navigation. The efficiency of the proposed system is demonstrated using accuracy of lane detection on straight and curve lanes covering frames per second and motor of the speed as parameters for assessment. Results show that the proposed system detects lane as entire lane detection, partial lane detection and no detection status on different settings.

This article investigates the effectiveness of feature extraction and selection techniques for enhancing the performance of classifiers accuracy in Type II Diabetes Mellitus (DM) detection using microarray gene data. To address the inherent high dimensionality of the data by employing three Feature Extraction (FE) methods are used namely, Short-Time Fourier Transform (STFT), Ridge Regression (RR), and Pearson Correlation Coefficient (PCC). To further refine the data, meta-heuristic algorithms like Bald Eagle Search Optimization (BESO) and Red Deer Optimization (RDO) are utilized for feature selection. The performance of seven classification techniques such as Non-Linear Regression - NLR, Linear Regression - LR, Gaussian Mixture Model - GMM, Expectation Maximization - EM, Logistic Regression - LoR, Softmax Discriminant Classifier - SDC, and Support Vector Machine with Radial Basis Function kernel - SVM-RBF are evaluated with and without feature selection. The analysis reveals that the combination of PCC with SVM-RBF achieved a promising accuracy of 92.85% even without feature selection. Notably, employing BESO with PCC and SVM-RBF maintained this high accuracy. However, the highest overall accuracy of 97.14% was achieved when RDO was used for feature selection alongside PCC and SVM-RBF. These findings highlight the potential of feature extraction and selection techniques, particularly RDO with PCC, in improving the accuracy of DM detection using microarray gene data.

Traditional computer vision techniques aim to extract meaningful information from images but often depend on manual feature engineering, making it difficult to handle complex real-world scenarios. Fractional Calculus (FC), which extends derivatives to non-integer orders, provides a flexible way to model systems with memory effects and long-term dependencies, making it a powerful tool for capturing fractional rates of variation. Recently, Neural Networks (NNs) have demonstrated remarkable capabilities in learning complex patterns directly from raw data, automating computer vision tasks and enhancing performance. Therefore, the use of fractional calculus in neural network-based computer vision is a powerful method to address existing challenges by effectively capturing complex spatial and temporal relationships in images and videos. This paper presents a survey of Fractional Calculus Neural Network-based (FC NN-based) computer vision techniques for denoising, enhancement, object detection, segmentation, restoration, and NN compression. The survey compiles existing FFC NN-based approaches, elucidates underlying concepts, and identifies open questions and research directions. By leveraging FC's properties, FC NN-based approaches offer a novel way to improve the robustness and efficiency of computer vision systems.

HIV early detection (CD4 counts ≥350 cells/μL) is correlated with higher life expectancy among people living with HIV (PLHIV). Several factors including physical, cultural, structural, and financial barriers have contributed to HIV late detection. This is the first-of-its-kind study on the regional differences in early detection of HIV within Tajikistan and any country in the Central Asia region. Utilizing Tajikistan Ministry of Health's national HIV data (N=10,700) spanning 2010 to 2023, we developed median regression models with median CD4 cell count as the outcome and as predictors: time (years), region, age, gender, and area (urban/rural status). Individuals younger than 19 years old are more likely to be detected early for HIV, whereas those older than 39 years are more prone to late detection. Given their active engagement in age-related and sexual activities, those older than 39 years face a higher risk of HIV transmission to others if diagnosed late. Therefore, they require further attention and targeted interventions. Females are detected earlier compared to their male counterparts, regardless of region of residence. Rural populations are detected earlier in most years compared to their urban counterparts. The COVID-19 pandemic accelerated HIV early detection in 2021 but most regions have returned to near pre-pandemic levels of detection in 2022 and 2023. These fluctuations in early detection may signify a shift in attention and resource allocation, potentially affecting efforts to address access gaps during the initial year of the COVID-19 pandemic. There are significant health inequities among different demographic and geographic groups which warrant further attention.

Underwater visual detection technology plays a pivotal role in fields such as marine exploration. With the increasing demand for underwater monitoring, the quest for efficient and reliable methods for underwater target recognition has become particularly significant. To address this requirement, this study developed an innovative underwater object detection architecture based on YOLOv8, named YOLOv8-MU, aimed at significantly enhancing detection accuracy.By integrating the LarK module proposed in UniRepLKNet to optimize the backbone network, YOLOv8-MU aims to achieve a larger receptive field without increasing the model’s depth. Further, this research introduces C2fSTR, an innovative method that combines Swin Transformer with the C2f module. Additionally, we have incorporated the SPPFCSPC_EMA module, which combines Cross-Stage Partial Fast Spatial Pyramid Pooling (SPPFCSPC) with attention mechanisms, significantly improving the detection accuracy and robustness of various biological targets. Moreover, by introducing a fusion block based on DAMO-YOLO into the neck of the model, we further enhanced the model’s capability in multi-scale feature extraction. Finally, the adoption of the MPDIoU loss function, designed around vertex distance, effectively tackles the challenges of localization accuracy and boundary clarity in underwater organism detection. Experimental results on the URPC2019 dataset demonstrate that the YOLOv8-MU model achieved an [email protected] of 78.4%, marking improvements of 5.6%, 1.1%, and 4.0% over YOLOv5s, YOLOv7, and YOLOv8n respectively, indicating the leading performance (SOTA) of this method. On the other hand, further evaluation on the URPC2020 dataset confirmed the generalization capability of the YOLOv8-MU architecture, with its [email protected] reaching 80.4%, surpassing various models including YOLOv5x and YOLOv8n, showcasing the wide applicability of our proposed improved model architecture.

Recently, a significant number of tools for detecting dangerous objects have been developed. Unfortunately, the performance offered by them is overestimated due to the poor quality of the datasets used (insufficiently numerous, contain items not strictly related to dangerous objects, insufficient range of presentation conditions). To fill in a gap in this area we have built an extensive dataset dedicated to detecting the objects most often used in various acts of breaching public security (baseball bat, gun, knife, machete, rifle). This collection contains images presenting the detected objects with different quality and under different environmental conditions. We believe that the results obtained from it are more reliable and give a better idea of the detection accuracy that can be achieved under real conditions. We used the Faster R-CNN with different backbone networks in the study. The best results were obtained for the ResNet152 backbone. The mAP value was 85%, while the AP level ranged from 80% to 91%, depending on the item detected. An average real-time detection speed was 11-13 FPS. Both the accuracy and speed of the Faster R-CNN model allow it to be recommended for use in public security monitoring systems aimed at detecting potentially dangerous objects.

Even with 100% certainty of a complete cure for breast cancer (BC), there is still a long way to go toward more efficient treatment. This requires sensitive and timely detection and accurate pre/post-clinical characterizations. Despite the availability of advanced diagnostic tools, many cancer patients lack access to efficient diagnostics that are both highly reliable and affordable. The fluorescence-based optical technique aims to make another significant leap forward in improving patient safety. It offers a convenient operation that reduces healthcare costs compared to visual examination tools (VETs). The primary and metastatic stages of BC consider different cancerous cell lines (MDAs), meaning the highest number of cells in this research (up to 300,000) represents the metastatic stages of BC, and 50,000 represents the primary level of BC developments have been studied based on fluorescence-enhanced photodynamic characterizations. The ability to characterize the fluorescence caused by MDA with 50,000 cells compared to the dominant radiation of MDA with 300,000 cells is emphatic proof of the high potential of fluorescence technique in timely BC detections, specifically before it spreads to the axillary lymph nodes. The specific cell numbers of 50,000 and 300,000 were chosen arbitrarily based on the cultivation of common biological limitations. Comparing the outcomes between 50,000 and 300,000 cells allows for evaluating the fluorescence technique’s diagnostic capability across various stages of breast cancer. This assessment provides valuable insights into the effectiveness of the fluorescence-based characterizing approach in detecting cancerous cells at different stages of the disease. Here, we have assessed fluorescence’s spectral shift and intensity difference as a diagnostic approach to distinguish between cancerous and normal breast cells. This study also presents a two-way structure of the 5-aminolevulinic acid (5-ALA) prodrug and Fluorescein Sodium (FS) effect in BC cell characterization from the perspective of photodynamical procedures and the detection side. 5-ALA induces an accumulation of protoporphyrin IX (PpIX) photosensitizer through a biosynthetic pathway, leading to red radiation of fluorescence measurements depending on different factors, such as temperature, incubation time, added glucose of the culturing medium, as well as photosynthesis processes. The presence and progression of breast cancer can be indicated by elevated levels of Reactive Oxygen Species (ROS), which are associated with the production of PpIX in cells following the administration of 5-ALA. Also, nicotinamide adenine dinucleotide (NADH) and flavin adenine dinucleotide (FAD) fluorophores are recognized as the main factors for fluorescence emissions at around 420-580 nm emission intervals. Considering the MDA’s high metastatic potential, the impact of 5-ALA on MDA’s cellular morphology and viability has been investigated. The molecular fluorophores are the primary probes to MDA’s cellular photodynamic considerations, allowing this widespread pre/post-clinical approach. The fluorescence signal reduction due to decreased cell viability and increased MDA’s cellular death rate after 24 hours of the 5-ALA-induced staining corresponds to the changes in lipid metabolism enzymes of MDAs cultured at different doses, which could be known as a cell death inducer function. Furthermore, statistical concerns have been studied using PCA multivariate component analysis to differentiate MDA cell lines administrated by 5-ALA.

Stress has various impacts on the health of human beings. Recent success in wearable sensor development, combined with advancements in deep learning to automatically detect features from raw data, opens several interesting applications related to detecting emotional states. Being able to accurately detect stress-related emotional arousal in an acute setting can positively impact the imminent health status of humans, i.e., through avoiding dangerous locations in an urban traffic setting. This work proposes an explainable deep learning methodology for the automatic detection of stress in physiological sensor data, recorded through a non-invasive wearable sensor device, the Empatica E4 wristband. We propose an Long-Short Term-Memory (LSTM) network, extended through a Deep Generative Ensemble of conditional GANs (LSTM DGE) , to deal with the low data regime of sparsely labeled sensor measurements. As explainability is often a main concern of deep learning models, we leverage Integrated Gradients (IG) to highlight the most essential features used by the model for prediction and to compare the results to state-of-the-art expert-based stress detection methodologies in terms of precision, recall and interpretability. Results show that our LSTM DGE outperforms the state-of-the-art algorithm by 3 percentage points in terms of recall, and 7.18 percentage points in terms of precision. More importantly, through the use of Integrated Gradients as a layer of explainability, we show that there is a strong overlap between model-derived stress features for electrodermal activity, and existing literature, which current state-of-the-art stress detection systems in medical research and psychology are based on.

This study discusses the application of deep learning technology in network intrusion detection systems (IDS) and focuses on a new model named CNN-Focal. First, through the review of traditional IDS technology, it analyzes its limitations in dealing with complex network traffic. Then, the design principle of the CNN-Focal model is described in detail, which uses threshold convolution and SoftMax multi-class classification technology to effectively improve abnormal traffic detection's accuracy and efficiency. The experimental results show that CNN-Focal performs well on the open data set, demonstrating the potential and advantages of its application in the natural network environment and providing a new perspective and method for further research of deep learning in the field of network security in the future.

Microarray gene expression analysis is a powerful technique used in cancer classification and research to identify and understand gene expression patterns that can differentiate between different cancer types, subtypes, and stages. However, microarray databases are highly redundant, inherently nonlinear, and noisy. Therefore, extracting meaningful information from such a huge database is a challenging one. The paper adopts the Fast Fourier Transform (FFT) and Mixture Model (MM) for dimensionality reduction and utilizes the Dragonfly optimization algorithm as the feature selection technique. The classifiers employed in this research are Nonlinear Regression, Naïve Bayes, Decision Tree, Random Forest and SVM (RBF). The classifiers' performances are analyzed with and without feature selection methods. Finally, Adaptive Moment Estimation (Adam) and Random Adaptive Moment Estimation (RanAdam) hyper-parameter tuning techniques are used as improvisation techniques for classifiers. The SVM (RBF) classifier with the Fast Fourier Transform Dimensionality Reduction method and Dragonfly feature selection achieved the highest accuracy of 98.343% with RanAdam hyperparameter tuning compared to other classifiers.

Statistical denoising techniques have been well studied, however many challenges still remain. This 9 article focuses on determining the least necessary number of significant samples in a statistical pop-10 ulation sampling, while minimizing the impact on the statistical detection of the fiber Bragg grating 11 (FBG) power spectra. This is achieved by using Bayesian decision making, small population statis-12 tics while excluding false detection. First, we investigate the impact of two-sided sliding window 13 technique applied near the cell under the test. The window size varies from 8 to 60 samples. The 14 considered maximum size of population sampling is connected with the number of wavelength 15 samples within the bandwidth of the symmetrical FBG power spectra peaks. Regarding the sym-16 metry of FBG power spectra peaks and the possible reduction of computational demands, we inves-17 tigate the impact of a one-sided sliding window. This enables to achieve less processing latency for 18 real time applications thus brings benefits to FBG-based fiber optical sensing. Both, two- and one-19 sided small population sampling techniques are then experimentally verified. Detection results are 20 analyzed in depth via calculating detection thresholds by using the small size window sliding along-21 side the waveband. From results achieved, it was found that the normality three-sigma rule does 22 not need to be followed while using the small population sampling. Experimental demonstrations 23 and analyses showed that the proposed concept of denoising and statistical threshold detection 24 based on the sliding window does not depend on the prior knowledge of probability distribution 25 functions of the FBG power spectra peaks level and background noise. We have also found that the 26 thresholds’ adaptability mainly depends on the statistical numerical characteristics of the above ad-27 ditive mixture of the FBG power spectra peaks signals and background noise.

The rapid evolution of technology has given rise to a connected world, where billions of devices interact seamlessly, forming what is known as the Internet of Things (IoT). While IoT offers incredible convenience and efficiency, it presents a significant challenge in cybersecurity and is characterized by various power, capacity, and computational process limitations. Machine learning techniques are employed within Intrusion Detection Systems (IDS) to enhance their capabilities in identifying and responding to security threats. The key features selected to improve IDS efficiency and reduce dataset size, thereby decreasing the time required for attack detection, are drawn from the extensive network dataset. This paper introduces an enhanced feature selection method designed to reduce the computational overhead on IoT resources while simultaneously strengthening intrusion detection capabilities within the IoT environment. Experimental results based on the InSDN dataset demonstrate that our proposed methodology achieves the highest accuracy with the fewest number of features and low computational cost. Specifically, we attain a 99.99% accuracy with 11 features and a computational time of 0.8599 seconds.

Fall detection systems play a crucial role in addressing the significant health concern of elderly falls, a leading cause of health deterioration and mortality. As the aging population grows and life expectancy increases, the development of accessible tools becomes vital for predicting and preventing falls, offering a practical and widely applicable solution in contrast to costly and expertise-dependent assessment tools. In contrast, due to the formidable challenges encountered, a comprehensive investigation into the comparative performance of standard ML models within this field still needs to be explored. This paper proposes a standard pipeline for pre-processing, training, and evaluating ML models for fall detection on the SisFall dataset. We conducted extensive experiments to evaluate the performance of various ML models for fall detection. The results validate the efficiency of the deep model in identifying the time windows in which a fall occurred. Among the deep models, the architecture, including a combination of convolutional neural networks and fully connected layers, outperforms the others by macro-averaged Precision, macro-averaged Recall, and macro-averaged F1-Score of 87.03\%, 86.83\%, and 86.93\%, respectively.

This paper presents a comprehensive study on applying machine learning (ML) techniques for real-time credit card fraud detection. With the increasing prevalence of credit card fraud, financial institutions face the challenge of detecting fraudulent transactions promptly and accurately. This research evaluates various ML models, including Logistic Regression, Decision Trees, Random Forests, XGBoost, and Deep Convolutional Neural Networks, for their effectiveness in identifying fraudulent activities. Utilizing PySpark for processing large-scale transaction data, the study highlights the importance of real-time analysis, adaptive learning, and handling imbalanced datasets. Key findings reveal that XgBoost, with its balance of accuracy and complexity, emerged as the most promising model, achieving an accuracy rate of 98% in detecting fraudulent transactions. The paper further discusses the potential of ensemble methods, graph-powered systems, and intelligent sampling in enhancing fraud detection capabilities. Challenges such as overfitting, data access, and the need for real-time analysis are addressed. Future research directions include implementing models in live transaction environments, expanding model applicability across financial platforms, and developing advanced anomaly detection methodologies. This study underscores the pivotal role of machine learning in safeguarding financial transactions against fraud, offering significant implications for consumers, financial institutions, and the broader financial ecosystem.

The autonomous displacement and displacement measurement functions of the maglev ruler are performed by the mover core. The stability of the four magnetic field in air gaps and the levitation position of the maglev ruler is one of the key factors for the stability of the thrust force on the power core, and it is also one of the key factors for ensuring the precision of the maglev ruler. There is cross-coupling between the two ends of the mover core of the maglev ruler, resulting in a strongly coupled, nonlinear, multi-input and multi-output system for the levitating system of the maglev ruler. This paper establishes a mathematical model for the levitating system of the maglev ruler, and designs a levitating control system for the maglev ruler based on ADRC control algorithm to achieve decoupling and disturbance suppression. Through simulation analysis and experimental testing of the levitating system with starting and disturbance, it is proved that the levitating control system of the maglev ruler has good dynamic characteristics, static characteristics, and robustness.

Despite significant advancements in Artificial Intelligence (AI) and Large Language Models (LLMs), detecting and mitigating bias remains a critical challenge, particularly within social media platforms like X (formerly Twitter) and in addressing cyberbullying present on them. This research investigates the effectiveness of leading LLMs in generating synthetic biased and cyberbullying data and evaluates the proficiency of Transformer AI models in detecting bias and cyberbullying within both authentic and synthetic contexts. The study involves semantic analysis and feature engineering on a dataset of over 48,000 sentences related to cyberbullying collected from Twitter (before it became X). Leveraging state-of-the-art LLMs such as ChatGPT-4o, Pi AI, Claude 3 Opus, and Gemini-1.5, synthetic biased, cyberbullying, and neutral data were generated to deepen the understanding of bias in human-generated data. AI models including DeBERTa, Longformer, BigBird, HateBERT, MobileBERT, DistilBERT, BERT, RoBERTa, ELECTRA, and XLNet were initially trained to classify Twitter cyberbullying data and subsequently fine-tuned, optimized, and quantized for multilabel classification (detecting both biases and cyberbullying). The study's outcomes include a prototype of a hybrid application that combines a Bias Data Detector and a Bias Data Generator, validated through extensive testing.

Detection of subsurface archaeological remains using a range of remote sensing methods, poses several challenges until today. Recent studies regarding the detection of archaeological proxies like those of cropmarks highlight the complexity of the phenomenon. In this work we present three different methods, and associated indices, for identifying stressed reflectance signatures indicating buried archaeological remains, based on a dataset of measured ground spectroradiometric reflectance. Several spectral profiles between the visible and near infrared part of the spectrum, were taken over a controlled environment in Cyprus during the 2011-2012 and are re-used in this study. The first two (spectral) methods are based on a suitable analysis of the spectral signatures in (1) the visible part of the spectrum, and in particular in the neighborhood of 570 nm, and (2) the red edge-near infrared part of the spectrum, in the neighborhood of 730 nm. Machine learning (decision trees) allows for the deduction of suitable wavelengths to focus on, in order to formulate the proposed indices and the associated classification criteria (decision boundaries) that can enhance the detection probability of stressed vegetation. Noise in the signal is taken into account by simulating reflectance signatures perturbed by white noise. Applying decision tree classification on the ensemble of simulations and basic statistical analysis we refine the formulation of the indices and criteria for the noisy signatures. The success rate of the proposed methods is over 90%. The third method rests on the estimation of vegetation/canopy reflectance parameters through inversion of the physical-based PROSAIL reflectance model and the associated classification through machine learning methods. The obtained results provide further insights into the formation of stress vegetation that occurred due to the presence of shallow buried archaeological remains, which are well aligned with physical-based models and existing empirical knowledge. To the best of the authors' knowledge, this is the first study demonstrating the usefulness of radiative transfer models such as PROSAIL for understanding the formation of cropmarks. Similar studies can support future research directions towards the development of regional remote sensing methods and algorithms if systematic observations are adequately dispersed in space and time.

While cycling presents environmental benefits and promotes a healthy lifestyle, the risks associated with overtaking maneuvers by motorized vehicles pose significant hindrances for many potential cyclists. Large-scale analysis of overtaking maneuvers could inform traffic researches and city planners on how to reduce these risks by better understanding these maneuvers. Drawing from the fields of sensor-based cycling research and of LiDAR-based traffic data sets, this paper provides a step towards addressing these safety concerns by introducing the Salzburg Bicycle 3d (SaBi3d) data set, consisting of LiDAR point clouds capturing car-to-bicycle overtaking maneuvers. The data set, collected using a LiDAR-equipped bicycle, facilitates detailed analysis of a large quantity of overtaking maneuvers without the need for manual annotation through enabling automatic labeling by a neural network. Additionally, a benchmark result for 3d object detection using a competitive neural network is provided as a baseline for future research. The SaBi3d data set is structured identically to the nuScenes data set, and therefore offers compatibility with numerous existing object detection systems. This work provides valuable resources for future researchers to better understand cycling infrastructure and mitigate risks, thus promoting cycling as a viable mode of transportation.

Few-shot object detection (FSOD) aims to address the challenge of requiring a substantial amount of annotations for training in conventional object detection, which is very labor-intensive. However, existing few-shot methods achieve high precision with the sacrifice of time-consuming for exhaustive fine-tuning, or take poor performance in novel-class adaptation. We presume the major reason is that the valuable correlation feature among different categories is insufficiently exploited, hindering the generalization of knowledge from base to novel categories for object detection. In this paper, we propose Few-Shot object detection via Correlation-RPN and Transformer Encoder-Decoder (CRTED), a novel training network to learn object-relevant features of inter-class correlation and intra-class compactness while suppressing object-agnostic features in the background with limited annotated samples. And we also introduce a 4-way tuple-contrast training strategy to positively activate the training progress of our object detector. Experiments over two few-shot benchmarks (Pascal VOC, MS-COCO) demonstrate that, our proposed CRTED without further fine-tuning can achieve comparable performance with current state-of-the-art fine-tuned works. The codes and pre-trained models will be released.

Fire is a serious security threat that can lead to casualties, property damage, and environmental damage. However, despite the availability of object detection algorithms, challenges persist in detecting fires and smoke. These challenges include slow convergence speed, poor performance in detecting small targets, and high computational cost limiting deployments. In this paper, Fire smoke and human detection based on ConvNeXt and Mixed encoder (FCM-DETR), an end-to-end object detection algorithm based on Deformable DETR, is proposed. Firstly, we introduce ConvNeXt to take the place of Resnet, which greatly reduces the amount of computation and improves the ability to extract irregular flame features. Secondly, to effectively process multi-scale features, the original encoder is decoupled into two modules. Then Mixed encoder, an innovative encoder structure, is proposed, resulting in excellent performance of multi-scale fire and smoke features fusion. What can’t be overlooked is that the encoder block is compatible with any DETR-based models. Finally, the convergence speed is accelerated and the mean of average precision is improved by applying a novel loss function called Powerful IoU v2. The experimental results indicate that our model achieves the best detection accuracy compared to other models, with mAP reaching 66.7%, mAPs and Accuracyfire achieving impressive 50.2% and 98.05%, respectively.

The complex network is an abstract modeing of complex systems in the real world, which plays an important role in analyzing the function of complex systems. Community detection is an important tool for analyzing network structure. In this paper, we propose a new community detection algorithm (RWBS) based on different seed nodes aims to understand the community structure of the network, which provides a new idea for the allocation of resources in the network. RWBS provides a new centrality metric ($MC$) to calculate node importance, which calculates the ranking of nodes as seed nodes. Furthermore, two algorithms are proposed for determining seed nodes on networks with and without ground-truth, respectively. We set the number of steps for the random walk to 6 according to the six degrees of separation theory to reduce the running time of the algorithm. Since some traditional community detection algorithms may detect smaller communities, e.g., two nodes become one community, this may make the resource allocation unreasonable. Therefore, modularity ($ Q $) is chosen as the optimization function to combine communities, which can improve the quality of detected communities. Final experimental results on real-world and synthetic networks show that the RWBS algorithm can effectively detect communities.

In emergency situations, every second counts for an ambulance navigating through traffic. Efficient use of traffic light systems can play a crucial role in minimizing response time. This paper introduces a novel automated Real-Time Ambulance in an Emergency Detector (RTAIAED). The proposed system uses special Lookout Stations (LS) suitably positioned at a certain distance from each involved traffic light (TL), to obtain timely and safe transitions to green lights as the Ambulance in an Emergency (AIAE) approaches. The RTAIAED is particularly pertinent on one-way roads, addressing the challenge of regulating the sequence of traffic-lights-signals so as to minimize the time needed to safely grant the green signal to the AIAE. The proposed solution leverages a part-based model made of elementary detectors for video analysis, specifically realized with a customized YOLOv8 model, and for audio analysis, thanks to an additional neural network based on Mel Frequency Cepstral Coefficients (MFCCs). This way, the RTAIAED ensures the accurate and robust identification of an AIAE heading towards a traffic light in time to ensure a green light thanks to the strategic positioning of the LS detectors. Extensive experiments demonstrate the robustness of the approach and its reliable application in real-world scenarios thanks to its predictions in real-time, showcasing the ability to detect AIAEs even in challenging conditions such as noisy environments, nighttime, or adverse weather conditions, provided a suitable-quality camera is appropriately positioned. The proposed system can also find application in traffic flow management.

Security has become a critical issue in today’s highly distributed and networked systems. Network intrusion detection systems (NIDSs), especially signature-based NIDSs, are being widely deployed in a distributed network environment with the purpose of defending against a variety of network attacks. Most of the commercially available NIDSs are software based and rely on pattern matching to extract the threat from network traffic. The increase in network speed and traffic may make existing algorithms to become a performance bottleneck. Therefore it is very necessary to develop faster and more efficient pattern matching algorithm in order to overcome the troubles on performance of NIDSs. Therefore, we propose a multi fusion pattern matching algorithm for Network Intrusion Detection Systems. The results obtained in percentages from the proposed fusion algorithm given better values in terms processing time in milliseconds than the existing algorithms when data English text are applied to evaluate the fusion performances.

With the development of educational technology, machine learning and deep learning provide technical support for traditional classroom observation assessment. However, in real classroom scenarios, the technique faces challenges such as lack of clarity of raw images, complexity of datasets, multi-target detection errors and complexity of character interactions.Based on the above problems, a student classroom behavior recognition network incorporating super-resolution and target detection is proposed. To cope with the problem of unclear original images in the classroom scenario, SRGAN (Super Resolution Generative Adversarial Network for Images) is used to improve the image resolution and thus the recognition accuracy.To address the dataset complexity and multi-targeting problems, feature extraction is optimized and multi-scale feature recognition is enhanced by introducing AKConv and LASK attention mechanisms into the Backbone module of YOLOv8s algorithm. To improve the character interaction complexity problem, the CBAM attention mechanism is integrated to enhance the recognition of important feature channels and spatial regions. Experiments show that it can detect six behaviors of students raising their hands, reading, writing, playing cell phones, looking down, and lying on the table in high-definition images. And the accuracy and robustness of this network are verified.Compared with small target detection algorithms such as Faster R-CNN, YOLOv5, and YOLOv8s, this network has better detection performance and can efficiently deal with the behavior recognition of multiple students.

The proliferation of fake news on social media platforms poses significant challenges to information authenticity and public trust. In this study, the impact of various word embedding techniques on the performance of both machine learning and deep learning models for fake news detection is systematically evaluated. The comprehensive TruthSeeker dataset, which spans over a decade of labeled news articles and social media posts, is utilized. Three popular word embedding techniques are analyzed: TF-IDF, Word2Vec, and FastText. These embeddings were tested with multiple machine learning classifiers, including logistic regression, naive Bayes, K-nearest neighbors, support vector machines, multilayer perceptrons, decision trees, and random forests, as well as deep learning models, specifically convolutional neural networks (CNNs). Among machine learning models, the Support Vector Machine (SVM) with TF-IDF achieved the highest accuracy of 99.03%, precision of 98.92%, recall of 99.15%, and F1 score of 99.03%. The Multilayer Perceptron (MLP) also demonstrated strong performance with Word2Vec, reaching an accuracy of 95.24%, precision of 94.95%, recall of 95.50%, and F1 score of 95.07%. For deep learning models, CNNs combined with TF-IDF embeddings showed outstanding results, achieving an accuracy of 98.99%, precision of 98.56%, recall of 99.48%, and F1 score of 99.02%. These results highlight the critical role of word embeddings in enhancing model accuracy and reliability in misinformation detection.

This work presents a Fault-Tolerant Model Predictive Control (FTMPC) algorithm applied to a simulation model for sewer networks. The control system will aim to prevent contaminated water from escaping the network, especially during periods of heavy rainfall, that can cause combined sewer overflows (CSO), while maintaining the flow of water towards the wastewater treatment plant (WWTP) as close as possible to its nominal value. The main purpose of the work is to preserve the operation of the predictive controller as much as possible, in accordance with its operational objectives, when there may be anomalies affecting the elements of the control system, mainly sensors and actuators. Comparing the results obtained considering various types of faults, with situations of normal controlled operation and with the behavior of the sewer network when no control is applied, will allow some conclusions to be drawn at the end.

Rapid and accurate detection of road cracks is of great significance for road health monitoring, but currently this work is mainly completed through manual site surveys. Low-altitude UAV remote sensing can obtain images with centimeter or even subcentimeter ground resolution, which provides a new efficient and economical approach for rapid crack detection. Nevertheless, crack detection networks face challenges such as edge blurring and misidentification due to the heterogeneity of road cracks and the complexity of the background. To address these issues, we propose a real-time edge reconstruction crack detection network (ERNet) which adopts multi-level information aggregation to reconstruct crack edges and improve the accuracy of segmentation between the target and background. To capture global dependencies across spatial and channel levels, we propose an efficient bilateral decomposed convolutional attention module (BDAM) that combines depth separable convolution and dilated convolution to capture global dependencies across spatial and channel levels. To enhance the accuracy of crack detection, we use a coordinate-based fusion module that integrates spatial, semantic, and edge reconstruction information. In addition, we propose an automatic measurement of crack information for extracting the crack trunk and its corresponding length and width. Experimental results demonstrate that our network achieves the best balance between accuracy and inference speed.

False news refers to false, fake, or misleading information presented as real news. In recent years, there has been a noticeable increase in false news on the Internet. The goal of this paper was to study the automatic detection of such false news using machine learning and natural language processing techniques and to determine which techniques work the most effectively. This article first studies what constitutes false news, and how it differs from other types of misleading information. We also study the results achieved by other researchers on the same topic. After building a foundation to understand false news, and the various ways of automatically detecting it, this article provides its own experiments. These experiments were done on four different datasets, one that was made just for this article, and using 10 different machine learning methods. The results of this article were satisfactory and provided answers to the original research questions set up at the beginning of this article. This article could determine from the experiments that passive-aggressive algorithms, support vector machines, and random forests are the most efficient methods for automatic false news detection. This article also concluded that more complex experiments, such as using multiple levels of identifying false news or detecting computer-generated false news, require more complex machine learning models.

Financial fraud poses a severe threat to the integrity of corporate financial statements and market efficiency. While auditors play a crucial role in detecting and preventing such fraud, the unique characteristics of this domain, including class imbalance, high misclassification costs, and am-biguous attributes, pose significant challenges. Traditional classification algorithms may not be optimally suited for this task. This paper proposes a novel stacking ensemble learning approach to enhanced detecting of Financial Statement Scam. By combining the strengths of multiple base learners, including Random Forest, XGBoost, and Gradient Boosting Decision Trees, our model leverages their diverse predictions to improve overall performance. A Logistic Regression me-ta-learner is employed to integrate the outputs of the base learners, capturing their collective wisdom while mitigating individual weaknesses. Extensive experimentation using a real-world dataset shown the superiority of our approach, achieving remarkable recall, precision, accuracy, and area under the curve metrics. The proposed model outperforms individual base learners, under-scoring the efficacy of ensemble learning in tackling the intricate problem of financial fraud detection. This research contributes to the creation of substantial and robust fraud detection systems, fostering trust and transparency in financial markets.

Detection and measurement of pain are crucial in designing effective pain management interventions to alleviate suffering and prevent functional decline in the medical domain. Traditional methods of pain identification rely on verbal patient reports or observable signs, which can be imprecise or subjective due to variability in patient self-assessments. Hence, there is a pressing need for automated methods to standardize pain measurement procedures. This paper proposes a novel facial expression-based automated pain assessment system as a behavioral indicator for pain evaluation. Our system integrates a fusion structure comprising Convolutional Neural Networks (CNNs) and Tensor-based extended Quantitative Expression Descriptors Analysis (TXQEDA). This innovative approach allows for joint learning of robust pain-related facial expression features from raw facial images, combining RGB appearance with shape-based latent representation. We extensively evaluated our proposed model using the UNBC-McMaster dataset for pain classification and intensity estimation. Our results demonstrate the efficacy of the proposed technique, achieving an impressive accuracy of 99.10% for pain level classification. Furthermore, our system excels in pain intensity estimation, showcasing its potential to provide precise and reliable assessments of pain severity. Overall, our research underscores the importance of automated pain assessment systems in improving patient care and clinical outcomes. By leveraging advanced deep learning techniques and fusion strategies, our proposed model offers a promising approach for standardizing pain measurement procedures and enhancing the quality of pain management interventions.

In today's digital landscape, cybersecurity has become a priority, with attacks becoming increasingly sophisticated. Traditional security approaches are no longer enough, requiring a more dynamic and advanced response. In this context, integrating Artificial Intelligence (AI) and Machine Learning (ML) appears key to addressing this growing threat. However, despite their high effectiveness, there is a developed problem: the integration of various data sources and technologies for comprehensive protection. This article presents an in-depth review of integrating Artificial Intelligence and Machine Learning in cybersecurity, focusing particularly on Open Extended Detection and Response (Open XDR) technology. The literature review in this article was meticulously conducted with specific criteria in mind. It focused on sourcing peer-reviewed journals, authoritative cybersecurity publications, and recent conferences that primarily concentrate on the advancements in AI and ML within the cybersecurity domain, particularly from 2018 to 2023. This timeframe was chosen to ensure the inclusion of the most current and relevant developments in Open XDR technology, Intrusion Detection Systems, Endpoint Detection and Response, and Security Information and Event Management systems. Additionally, the review gave special attention to studies and reports highlighting practical implementations and real-world applications of these technologies, thereby ensuring a comprehensive understanding of their impact and effectiveness in enhancing cybersecurity resilience. The methodology used is a detailed literature review, examining how various cybersecurity components interact and function. These components include Intrusion Detection Systems (IDS), which monitor networks for malicious activities; Endpoint Detection and Response (EDR), which focuses on detecting and investigating security incidents on endpoints; and Security Information and Event Management (SIEM), systems that provide real-time analysis of security alerts. The review also considers the role of Active Directory, a directory service for Windows domain networks, and the process of log forwarding, where log files are transmitted to a central server for analysis, in the context of AI and ML. The paper delves into the development of AI and ML, underscoring their roles in cybersecurity for advanced data processing, pattern recognition, and predicting threats. It explores both supervised (where the model is trained on labeled data) and unsupervised learning (where the model learns from unlabeled data) in ML, and how these techniques bolster cybersecurity measures. The article highlights the significance of Open XDR as a critical innovation that integrates data from multiple sources for comprehensive security analysis. Further, the review discusses how the integration of AI and ML into various cybersecurity tools, such as IDS, EDR, and SIEM, augments capabilities in threat detection and response. It addresses the challenges and opportunities that AI and ML present in the cybersecurity domain, focusing on ethical issues, data privacy concerns, and the necessity for ongoing professional development in this rapidly advancing field. The paper concludes by affirming the effectiveness of merging AI and ML with these cybersecurity tools within the Open XDR framework.

Large datasets play a crucial role in the progression of surgical robotics, facilitating advancements in the fields of surgical task recognition and automation. Moreover, public datasets enable the comparative analysis of various algorithms and methodologies, thereby assessing their effectiveness and performance. The ROSMA (Robotics Surgical Maneuvers) dataset provides 206 trials of common surgical training tasks performed with the da Vinci Research Kit (dVRK). In this work, we extend ROSMA dataset with two annotated subsets: ROSMAT24, which contains bounding box annotations for instrument detection, and ROSMAG40, which contains high and low level gesture annotations. We propose an annotation method that provides independent labels for the right-handed tool and the left-handed tool. For instrument identification, we validate our proposal with a YOLOv4 model in two experimental scenarios. We demonstrate the generalization capabilities of the network to detect instruments in unseen scenarios. For gesture segmentation, we propose two label categories: high-level annotations that describe gestures at a maneuvers level, and low-level annotations that describe gestures at a fine-grain level. To validate this proposal, we have designed a recurrent neural network based on a bidirectional long-short term memory layer. We present results for four cross-validation experimental setups, reaching up to a 77.35% mAP.

The development of a rapid, sensitive, specific method for detecting foodborne pathogens is paramount in supplying safe food to enhance public health safety. Despite the significant improvement in pathogen detection methods, key issues are still associated with rapid methods, such as distinguishing living cells from dead, pathogenic potential or health risk of the analyte at the time of consumption, detection limit, and sample-to-result. The mammalian cell-based assays analyze pathogens' interaction with the host cells and are responsive to only live pathogens or active toxins. In this study, a human embryonic kidney (HEK293) cell line expressing Toll-Like Receptor 5 (TLR-5) and chromogenic reporter system (HEK dual hTLR5) was used for the detection of viable Salmonella in a 96-well tissue culture plate. This cell line responds to low concentrations of TLR5 agonist flagellin. Stimulation of TLR5 ligand activates nuclear factor-kB (NF-κB) - linked alkaline phosphatase (AP-1) signaling cascade inducing the production of secreted embryonic alkaline phosphatase (SEAP). With the addition of a ρ-nitrophenyl phosphate as a substrate, a colored end product representing a positive signal is quantified. The assay's specificity was validated with the top 20 Salmonella enterica serovars and 19 non-Salmonella spp. The performance of the assay was also validated with spiked food samples. The total detection time (sample-to-result), including shortened pre-enrichment (4 h) and selective enrichment (4 h) steps with artificially inoculated outbreak-implicated food samples (chicken, peanut kernel, peanut butter, black pepper, mayonnaise, and peach) was 15 h when inoculated at 1–100 CFU/25 g sample. These results show the potential of HEK-DualTM hTLR5 cell-based functional biosensors for rapid screening of Salmonella.

In response to the burgeoning global demand for premium agricultural products, particularly within the competitive nut market, this paper introduces an innovative methodology aimed at enhancing the grading process for almonds and their shells. Leveraging state-of-the-art Deep Convolutional Neural Networks (CNNs), specifically the AlmondNet-20 architecture, our study achieves exceptional accuracy exceeding 99%, facilitated by the utilization of a 20-layer CNN model. To bolster robustness in differentiating between almonds and shells, data augmentation techniques are employed, ensuring the reliability and accuracy of our classification system. Our model, meticulously trained over 1000 epochs, demonstrates remarkable performance, boasting an accuracy rate of 99% alongside a minimal loss function of 0.0567. Rigorous evaluation through test datasets further validates the efficacy of our approach, revealing impeccable precision, recall, and F1-score metrics for almond detection. Beyond its technical prowess, this advanced classification system offers tangible benefits to both industry experts and non-specialists alike, ensuring globally reliable almond classification. The application of deep learning algorithms, as showcased in our study, not only enhances grading accuracy but also presents opportunities for product patents, thereby contributing to the economic value of our nation. Through the adoption of cutting-edge technologies such as the AlmondNet-20 model, we pave the way for future advancements in agricultural product classification, ultimately enriching global trade and economic prosperity.

The presence of cloud and cloud shadows in satellite images affects the accuracy of land cover classification and object detection. Therefore, their existence is considered noise and a target for removal by many researchers. This paper focuses on precise cloud shadow detection from ortho-rectified images. The directions of cloud shadows in original satellite images are determined by the sensor and sun illumination directions. However, those in ortho-rectified images have not been studied explicitly. This study proposes that the directions of cloud shadow in ortho-rectified images are also determined by both sensor and sun directions. This is because relief displacements due to cloud height above ground are not corrected through ortho-rectification processes. This study also proposed to detect cloud shadows by shifting a bounding box of a cloud region through the cloud shadow direction and identifying dark pixels within the shifted box. This study utilized Rapideye images acquired in various viewing angles and viewing directions. The proposed method improved the accuracy of estimating cloud shadow directions and extracting cloud shadow regions greatly. The outcomes of this study are expected to be utilized as precise cloud shadow detection and correction. As a future work, there is a need to enhance accuracy of cloud shadow detection through post-processing methods such as watershed algorithm. Additionally, an interesting future study would be to check whether the findings of this study are applicable to other objects in ortho- or unmanned aerial vehicle (UAV) images with uncorrected relief displacements, such as high-rise buildings.

Growth monitoring of crops is a crucial aspect of precision agriculture, essential for optimal yield prediction and resource allocation. Traditional crop growth monitoring methods are labor-intensive and prone to errors. This study introduces an automated segmentation pipeline utilizing multi-date aerial images and ortho-mosaics to monitor the growth of cauliflower crops (Brassica Oleracea var. Botrytis) using an object-based image analysis approach. The methodology employs YOLOv8, Grounding Detection Transformer with Improved Denoising Anchor Boxes (DINO), and the Segment Anything Model (SAM) for automatic annotation and segmentation. The YOLOv8 model was trained using aerial image datasets, which then facilitated the training of the Grounded Segment Anything Model framework. This approach generated automatic annotations and segmentation masks, classifying crop rows for temporal monitoring and growth estimation. The study’s findings utilized a multi-modal monitoring approach to highlight the efficiency of this automated system in providing accurate crop growth analysis, promoting informed decision-making in crop management and sustainable agricultural practices. Results indicate consistent and comparable growth patterns between aerial images and ortho-mosaics, with significant periods of rapid expansion and minor fluctuations over time. The results also indicated a correlation between the both the time and method of observation which paves a future possibility of integration of such techniques aimed at increasing the accuracy in crop growth monitoring based on automatically derived temporal crop row segmentation masks.

Of the 100,000 railroad bridges in the United States, 50% are over 100 years old. Many of these bridges do not meet the minimum vertical clearance standards, making them susceptible to impacts from over-height vehicles. Impacts can cause structural damage and unwanted disruption to railroad bridge services; rapid notification of the railroad authorities is crucial to ensure that the bridges are safe for continued use and to affect timely repairs. Therefore, researchers have developed approaches to identify these impacts to railroad bridges. Some recent approaches use machine learning to more effectively identify impacts from the sensor data. Typically, the collected sensor data is transmitted to a central location for processing. However, the challenge with this centralized approach is that the transfer of data to a central location can take considerable time, which is undesirable for time-sensitive events like impact detection that require rapid assessment and response to potential damage. To address the challenges posed by the centralized approach, this study develops a framework for edge implementation of machine learning predictions on wireless smart sensors. Wireless sensors are used because of their ease of installation and lower costs compared to their wired counterparts. The framework is implemented on the Xnode wireless smart sensor platform, thus bringing artificial intelligence models directly to the sensor nodes and eliminating the need to transfer data to a central location for processing. This framework is demonstrated using data obtained from events on a railroad bridge near Chicago; results illustrate the efficacy of the proposed edge computing framework for such time-sensitive structural health monitoring applications.

Gastric cancer has become a serious worldwide health concern, emphasizing the cru-cial importance of early diagnosis measures to improve patient outcomes. While tradi-tional histological image analysis is regarded as the clinical gold standard, it is la-bor-intensive and manual. Recognizing this problem, there has been a rise in interest in using computer-aided diagnostics tools to help pathologists with their diagnostic ef-forts. In particular, deep learning (DL) has emerged as a promising solution in this sector. However, current DL models are still restricted in their ability to extract exten-sive visual characteristics for correct categorization. To tackle this limitation, this study proposes the use of ensemble models, which incorporate the capabilities of sev-eral deep learning architectures and use aggregate knowledge of many models to im-prove classification performance, allowing for more accurate and efficient gastric cancer detection. To see how well these proposed models performed, this study com-pared them to other works, all of which were based on the Gastric Histopathology Sub-size Images Database, a publicly available dataset for gastric cancer. This research demonstrated that the ensemble models achieved high detection accuracy across all sub-databases, with an average accuracy exceeding 99%. Specifically, ResNet50, VGGNet, and ResNet34 performed better than EfficientNet and VitNet. For the 80 × 80 pixel sub-database, ResNet34 exhibited an accuracy of approximately 93%, VGGNet achieved 94%, and the ensemble model excelled with 99%. In the 120 × 120 pixel sub-database, the ensemble model showed 99% accuracy, VGGNet 97%, and ResNet50 approximately 97%. For the 160 × 160 pixel sub-database, the ensemble model again achieved 99% accuracy, VGGNet 98%, ResNet50 98%, and EfficientNet 92%, high-lighting the ensemble model's superior performance across all resolutions. Overall, the ensemble model consistently provided an accuracy of 99% across the three sub-pixel categories. These findings showed that ensemble models may successfully detect criti-cal characteristics from smaller patches and achieve high performance. The findings will help pathologists diagnose gastric cancer using histopathological images, leading to earlier identification and higher patient survival rates.

The proliferation of Internet of Things (IoT) devices on Linux platforms has heightened concerns regarding vulnerability to malware attacks. This paper introduces a novel approach to investigating the behavior of Linux IoT malware by examining syscalls and library syscall wrappers extracted through static analysis of binaries, as opposed to the conventional method of using dynamic analysis for syscall extraction. We rank and categorize Linux system calls based on their security significance, focusing on understanding malware intent without execution.Feature analysis of the assigned syscall categories and risk ranking is conducted with statistical tests to validate their effectiveness and reliability in differentiating between malware and benign binaries. Our findings demonstrate that potential threats can be reliably identified with an F1 score of 96.86%, solely by analyzing syscalls and library syscall wrappers. This method can augment traditional static analysis, providing an effective preemptive measure to enhance Linux malware analysis. This research highlights the importance of static analysis in strengthening IoT systems against emerging malware threats.

This work presents a methodology for the generation and classification of phase resolved partial discharges (PRPD) patterns based on the use of a printed UHF monopole antenna and signal con-ditioning circuit to reduce hardware requirements. For this purpose, envelope detection technique was applied. In addition, test objects such as a hydrogenerator bar, dielectric disks with internal cavities in an oil cell, potential transformer, and tip-tip electrodes immersed in oil were used to generate partial discharge (PD) signals. To detect and classify partial discharges, the standard IEC 60270 (2000) method was used as reference. After acquisition of conditioned UHF signals, digital signal filtering threshold technique was used, and peaks of partial discharges envelopes pulses were extracted. Feature selection techniques were used to classify the discharges and choose the best features to train machine learning algorithms, such as multilayer perceptron, support vector ma-chine and decision tree. Accuracies greater than 84% were met, revealing the classification potential of the methodology proposed in this work.

Tunnel fires are generally detected using various sensors, including measuring temperature, CO concentration, and smoke concentration. To address the ambiguity and inconsistency in multi-sensor data, this paper proposes a tunnel fire detection method based on an improved Dempster-Shafer (DS) evidence theory for multi-sensor data fusion. To solve the problem of evidence conflict in DS theory, a two-level multi-sensor data fusion framework is adopted. The first level of fusion involves feature fusion of the same type of sensor data, removing ambiguous data to obtain characteristic data, and calculating the basic probability assignment (BPA) function through the feature interval. The second level fusion derives basic probability numbers from the BPA, calculates the degree of evidence conflict, normalizes to obtain the relative conflict degree, and optimizes the BPA using the trust coefficient. The classical DS evidence theory is then used to integrate and obtain the probability of tunnel fire occurrence. Different heat release rates, tunnel wind speeds, and fire locations are set, forming four fire scenarios. Sensor monitoring data under each simulation condition are extracted and fused using the improved DS evidence theory. The results show that the proposed improved DS evidence theory method detects the probability of fire occurrence in the four scenarios as 67.5%, 71.0%, 82.8%, and 83.5%, respectively, and identifies fire occurrence in approximately 2.4 seconds, an improvement of 64.7% to 70% over traditional methods. This demonstrates the feasibility and superiority of the proposed method, highlighting its significant importance in ensuring personnel safety.

In addressing the multifaceted problem of multiple-input multiple-output (MIMO) detection in wireless communication systems, this work highlights the pressing need for enhanced detection reliability under variable channel conditions and MIMO antenna configurations. We propose a novel method that sets a new standard for deep unfolding approaches to MIMO detection by integrating the iterative conjugate gradient method with Tikhonov regularization, combining the adaptability of modern deep learning techniques with the robustness of classical regularization. Unlike conventional techniques, our strategy treats the regularization parameter of Tikhonov regularization as well as step size values and search direction coefficients of conjugate gradient (CG) method as trainable parameters within the deep learning framework, allowing for dynamic modification according to channel conditions and MIMO antenna configurations. Detection performance is significantly improved by the proposed approach in variety of conditions. In different MIMO settings, the suggested method consistently shows better bit error rate (BER) and normalized minimum mean square error (NMSE) performance. Across a range of MIMO configurations and channel conditions, the proposed method exhibits significantly lower BER and NMSE values than well-known techniques such as DetNet and CG. The proposed method has superior performance over CG and other model-oriented methods, especially in small number of iterations. Consequently, the simulation results demonstrate the flexibility of the proposed approach, making it a viable choice for MIMO systems with a range of antenna configurations and different channnel conditions.

In this paper we develop and explore a methodical pipeline for swimmer localisation in outdoor environments. The developed framework is intended to be used for enhancing swimmer safety. A main issue we deal with by the proposed approach is the lack of real world training data in such outdoor environments. Natural lighting changes, dynamic water textures and possibly barely visible swimming persons are key elements to approach. We account for these difficulties by adopting an effective background removal technique with available training data. This allows us to edit swimmers into natural environment backgrounds for the use in subsequent image augmentation. We created 17 training datasets with real images, synthetic images and a mixture of both to investigate different aspects and characteristics of the proposed approach. The datasets are used to train a YOLO architecture for the possible future application in real-time detection. The trained framework is then tested and evaluated on outdoor environment imagery acquired by a safety drone to investigate and confirm the usefulness for outdoor swimmer localisation.

Real-time object detection remains a pivotal topic in the realm of computer vision. Balancing the accuracy and speed of object detectors poses a formidable challenge for both academic researchers and industry practitioners. While recent transformer-based models have showncase the prowess of the attention mechanism, delivering remarkable performance improvements over CNNs, their computational demands can hinder their effectiveness in real-time detection tasks. In this paper, we elect to use YOLOX as our robust starting point and introduce a series of effective enhancements, culminating in a new high-performance detector named YOLOAX. To further exploit the power of the attention mechanism, we devise multi-dimensional attention-based modules that activate CNNs, emphasizing the most pertinent regions and bolstering the capacity to learn the most informative image representations from feature maps. Moreover, we introduce a novel leading label assignment strategy called STA, along with a groundbreaking loss function named GEIOU Loss, to further refine our detector's performance. We provide extensive ablation studies on the COCO and PASCAL VOC 2012 detection datasets to validate our proposed methods. Remarkably, our YOLOAX is trained solely on the MS-COCO dataset from scratch, without leveraging any prior knowledge, which achieves an impressive 54.2% AP on the COCO 2017 test set while maintaining a real-time speed of 72.3 fps, surpassing YOLOX by a significant margin of 3.0% AP. Our source code and pre-trained models are openly accessible at https://github.com/KejianXu/yoloax.

The Internet of Things (IoT) has a growing presence in society's daily lives. These lightweight devices can be easily deployed and maintained, enabling extensive adoption in different environments. Furthermore, one of the most promising areas for using IoT devices is healthcare, comprising devices referred to as the Internet of Medical Things (IoMT). Several examples of healthcare services are supported by IoMT devices, e.g., continuous health monitoring. Conversely, there is an increasing concern with the cybersecurity aspects of these devices, and several attacks against IoT infrastructures have been launched in the past few years. These cybersecurity concerns also apply to healthcare applications, where the tradeoff between the benefits and security of IoMT devices must be observed. Given the complexity and amount of data IoMT network traffic generates, advanced methods become especially useful in these environments. Although Machine Learning (ML) brings various techniques and solutions to improve cyberattack detection, prevention, and mitigation, essential features are not addressed in the current state-of-the-art benchmark dataset contributions. Thereupon, the main goal of this research is to propose a realistic benchmark dataset to enable the development and evaluation of IoMT security solutions. In order to accomplish this, 18 attacks were executed against an IoMT testbed composed of 40 IoMT devices (25 real devices and 15 simulated devices), considering the plurality of protocols used in healthcare (e.g., Wi-Fi, MQTT, and Bluetooth). These attacks are categorized into five classes: DDoS, DoS, Recon, MQTT, and spoofing. This effort aims to establish a baseline complementary to the state-of-the-art contributions. The outcome supports researchers in investigating and developing new solutions to make healthcare systems more secure using different mechanisms (e.g., machine learning - ML). This research goes beyond merely conducting attacks on IoMT devices. We also attempt to capture the lifecycle of these devices in different vital phases, from the moment they join the network until they leave, which is called profiling. Profiling allows the different classifiers to identify anomalies of each device individually in the healthcare network. The CICIoMT2024 dataset has been published on CIC's dataset page, making it available for other researchers to use.

With the increasing use of data-driven modeling methods new approaches to complex problems in the field of wind energy can be addressed. Topics reviewed through literature include wake modeling, performance monitoring and controls applications, condition monitoring and fault detection, and other data-driven research. Literature shows the advantages of data-driven methods reduce computational expense or complexity, particularly in the cases of wake modeling and controls, as well as various data-driven methodologies’ aptitude for predictive modeling and classification, as in the cases of fault detection and diagnosis. Significant work exists for fault detection while less work is found for controls applications. A methodology for creating data-driven wind turbine models for arbitrary performance parameters is proposed. Results are presented utilizing the methodology to create wind turbine models relating active adaptive twist to steady-state rotor thrust as a performance parameter of interest. Resulting models are evaluated by comparing root-mean-square-error (RMSE) on both the training and validation data sets, with Gaussian Process Regression (GPR), deemed an accurate model for this application. The resulting model undergoes particle swarm optimization to determine the optimal aerostructure twist shape at a given wind speed with respect to the modeled performance parameter, aerodynamic thrust load. The optimization process shows an improvement of 3.15% in thrust loading for the 10 MW reference turbine, and 2.66% for the 15 MW reference turbine.

In light of the critical and emerging global challenge posed by the increasing frequency and severity of heatwaves in recent decades, this study demonstrates a practical and robust workflow that bridges the gap in understanding how the effective integration of remote sensing data with ground-based observations can facilitate a comprehensive assessment of spatiotemporal heatwave patterns and trends in the central region of Thailand. This research transcends traditional methods, utilizing air temperature (Tair) and satellite-derived land surface temperature (LST) data from 1981 to 2019. Results exhibit that an analysis of Tair indicates increasing daytime heatwaves in peri-urban areas, with significant trends in heatwave number (HWN), heatwave frequency (HWF), and heatwave duration (HWD), heatwave amplitude (HWA) rises in urban settings while nighttime heatwaves significantly intensify in rural locales. Notably, LST trends reveal varied patterns, with peri-urban areas showing marked daytime increases in heatwave magnitude (HWM), HWA, and HWF. Correlation analysis (p=0.05) highlights strong daytime associations between Tair and LST in rural (HWN, HWF, HWD, r>0.90) and peri-urban (HWM, HWA, r>0.65) regions. Overall, this study advances approaches to the adaptable measurement and spatial-temporal pattern of heatwave-related risk areas, providing insights for decision-makers to develop sustainable practices and strategies for climate change adaptation.

The device and method in accordance with the constant pressure regulation of micro droplet PCR in microfluidic chips are developed to explore the optimization way for the micro droplet movement, fragmentation, and bubble generation in microfluidic chips in existing digital PCR technology. In the developed device, an air source device is adopted to regulate the pressure in the chip, such that micro droplet generation and PCR amplification without bubbles can be achieved. In 3 min, the sample in 20ul will be distributed into nearly 50,000 water-in-oil droplets exhibiting a diameter of about 87μm, and the micro droplet will be subjected to a close arrangement in the chip without air bubbles. The device and chip are adopted to quantitatively detect the human gene . As indicated by the experimental results, a good linear relationship exists between detection signal and DNA concentration ranging from 101~ 105 copies /μL (R2=0.999). The micro droplet PCR devices based on constant pressure regulation chips exhibit a wide variety of advantages (e.g., achieving high pollution resistance, micro droplet fragmentation and integration avoidance, reducing human interference, and standardizing results). Thus, micro droplet PCR devices based on constant pressure regulation chips have promising applications in nucleic acid quantification.

Recently, there has been considerable research on deepfake detection. However, most existing methods face challenges in adapting to the advancements in new generative models within unknown domains. In addition, the emergence of new generative models capable of producing and editing high-quality images, such as Diffusion, Consistency, and LCM, poses a challenge for traditional deepfake training models. These advancements highlight the need for adapting and evolving existing deepfake detection techniques to effectively counter the threats posed by sophisticated image manipulation technologies. In this paper, our objective is to detect deepfake videos in unknown domains using unlabeled data. Specifically, our proposed approach employs Meta Pseudo Labels (MPL) with Supervised Contrastive Learning, so-called SupCon-MPL, allowing the model to be trained on unlabeled images. MPL involves the simultaneous training of both a Teacher model and a Student model, where the Teacher model generates Pseudo Labels utilized to train the Student model. This method aims to enhance the adaptability and robustness of deepfake detection systems against emerging unknown domains. Supervised Contrastive Learning utilizes labels to compare samples within similar classes more intensively, while encouraging greater distinction from samples in dissimilar classes. This facilitates the learning of features in a diverse set of deepfake images by the model, consequently contributing to the performance of deepfake detection in unknown domains. When utilizing the ResNet50 model as the backbone, SupCon-MPL exhibited an improvement of 1.58% in accuracy compared to traditional MPL in known domain detection. Moreover, in the same generation of unknown domain detection, there was a 1.32% accuracy enhancement, while in the detection of post-generation unknown domains, there was an 8.74% increase in accuracy.

Artificial Intelligence (AI) is revolutionizing head and neck cancer (HNC) care by providing innovative tools that enhance diagnostic accuracy and personalize treatment strategies. This review highlights the advancements in AI technologies, including deep learning and natural language processing, and their applications in HNC. The integration of AI with imaging techniques, genomics, and electronic health records is explored, emphasizing its role in early detection, biomarker discovery, and treatment planning. Despite noticeable progress, challenges such as data quality, algorithmic bias, and the need for interdisciplinary collaboration remain. Emerging innovations like explainable AI, AI-powered robotics, and real-time monitoring systems are poised to further advance the field. Addressing these challenges and fostering collaboration among AI experts, clinicians, and researchers is crucial for developing equitable and effective AI applications. The future of AI in HNC holds significant promise, offering potential breakthroughs in diagnostics, personalized therapies, and improved patient outcomes.

Fire detection from satellite sensors relies on an accurate estimation of the unperturbed state of a target pixel, from which an anomaly can be isolated. Methods for estimating the radiation budget of a pixel without fire depend upon training data derived from the location's recent history of brightness temperature variation over the diurnal cycle, which can be vulnerable to cloud contamination and the effects of weather. This study proposes a new method that utilises the common solar budget found at a given latitude in conjunction with an area's local solar time to aggregate a broad-area training dataset, which can be used to model the expected diurnal temperature cycle of a location. This training data is then used in a temperature fitting process with the measured brightness temperatures in a pixel, and compared to pixel-derived training data and contextual methods of background temperature determination. Results of this study show similar accuracy between clear-sky medium wave infrared upwelling radiation and the diurnal temperature cycle estimation compared to previous methods, with demonstrable improvements in processing time and training data availability. This method can be used in conjunction with brightness temperature thresholds to provide a baseline for upwelling radiation, from which positive thermal anomalies such as fire can be isolated.

This article presents an Internet of Things (IoT)-enabled system meticulously engineered for comprehensive environmental assessment, encompassing air quality, temperature, and humidity. Simultaneously, the devised system integrates cutting-edge MQ6 sensors, adept at detecting ambient gases, with a primary focus on monitoring Liquefied Petroleum Gas (LPG) and Butane gases concentrations in the atmosphere. This distinctive feature not only facilitates the identifi-cation of hazardous gases leakages, particularly common in household LPG usage, but also empowers preemptive measures. The environmental parameters, air quality, temperature, and humidity, are precisely gauged through the utilization of MQ135 and DHT11 sensors. The con-trolling and managing of the hardware component of the system (i.e., device) is assured by an Arduino UNO, that orchestrates the collection of sensor data, transmitting it seamlessly to both the ThingSpeak cloud and a Convolutional Neural Network (CNN) model. By interfacing with ThingSpeak, the device ensures the systematic storage and visualization of environmental data, streamlining monitoring and analysis processes. In the event of perilous gases detection, the system promptly triggers an alarm, concurrently dispatching a real-time notification to the device owner via the integrated If-This-Then-That (IFTTT) application programming interface (API). To further enrich environmental awareness, an embedded microphone captures audio recordings, subsequently processed by a CNN-based deep learning model. This model employs advanced techniques such as spectrogram generation, Convolution2D, and MaxPooling2D within a Se-quential Model architecture, to attain elevated levels of accuracy, furnishing insightful interpre-tations of ongoing activities within the environment.

Biases in Artificial Intelligence (AI) or Machine Learning (ML) systems due to skewed datasets problematise the application of prediction models in practice. Representation bias is a prevalent form of bias found in the majority of datasets. This bias arises when training data inadequately represents certain segments of the data space, resulting in poor generalisation of prediction models. Despite AI practitioners employing various methods to mitigate representation bias, their effectiveness is often limited due to a lack of thorough domain knowledge. To address this limitation, this paper introduces human-in-the-loop interaction approaches for representation debiasing of generated data involving domain experts. Our work advocates for a controlled data generation process involving domain experts to effectively mitigate the effects of representation bias. We argue that domain experts can leverage their expertise to assess how representation bias affects prediction models. Moreover, our interaction approaches can facilitate domain experts in steering data augmentation algorithms to produce debiased augmented data and validate or refine the generated samples to reduce representation bias. We also discuss how these approaches can be leveraged for designing and developing user-centred AI systems to mitigate the impact of representation bias through effective collaboration between domain experts and AI.

Deep Neural Network (DNN) inference demands substantial computing power, resulting in significant energy consumption. A large number of negative output activations in convolution layers are rendered zero due to the invocation of the ReLU activation function. This results in a substantial number of unnecessary computations that consume significant amounts of energy. This paper presents ECHO: Energy-efficient Computation Harnessing Onilne Arithmetic - A MSDF-based accelerator for DNN inference, designed for computation pruning, utilizing an unconventional arithmetic paradigm known as online/ most-significant digit first (MSDF) arithmetic which performs computations in a digit-serial manner. The MSDF digital serial computation of online arithmetic enables overlapped computation of successive operations leading to substantial performance improvements. The online arithmetic, coupled with a negative output detection scheme, facilitates early and precise recognition of negative outputs. This, in turn, allows for the timely termination of unnecessary computations, resulting in a reduction of energy consumption. The implemented design has been realized on the Xilinx Virtex-7 VU3P FPGA and subjected to a comprehensive evaluation through a rigorous comparative analysis involving widely used performance metrics. Experimental results demonstrate promising power and throughput improvements compared to contemporary methods. In particular, the proposed design achieved an average improvement in power consumption of up to 81%, 82.9%, and 40.6% for VGG-16, ResNet-18, and ResNet-50 workloads compared to the conventional bit-serial design, respectively. Furthermore, significant average speedups of 2.39×, 2.6×, and 2.42 were observed when comparing the proposed design to conventional bit-serial designs for VGG-16, ResNet-18, and ResNet-50 models respectively.

Foodborne illnesses can be infectious and dangerous and most of them are caused by bacteria. Some types of common food-related bacteria exist widely in nature and pose a serious threat to both humans and animals, and can cause poisoning, diseases, disabilities and even death. Rapid, reliable and cost-effective methods for bacteria detection are of paramount importance in food-safety and environmental monitoring. Polymerase chain reaction (PCR), lateral flow immunochromatographic assay (LFIA) and electrochemical method have been widely used in food-safety and environmental monitoring. In this paper, the recent developments (2013-2023) covering PCR, LFIA and electrochemical method for various bacteria detection (Salmonella, Listeria, Campylobacter, Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli)) considering different foods types, analytical performances and the reported limit of detection (LOD) are discussed. It is found that the bacteria type and food sample type contribute significantly to the analytical performance and LOD. Detection by LFIA has higher average LOD (24 CFU/ml) than detection by electrochemical method (12 CFU/ml) and PCR (6 CFU/ml). Salmonella and E. coli in the Pseudomonadota domain usually have low LODs. LODs are usually lower for detection in fish and eggs. LFIA with gold and iron nanoparticles are prominent in the majority of articles of 26 CFU/ml and 12 CFU/ml respectively. Electrochemical method reveals that the average LOD is highest for cyclic voltammetry (CV) of 18 CFU/ml), followed by electrochemical impedance spectroscopy (EIS) of 12 CFU/ml and differential pulse voltammetry (DPV) of 8 CFU/ml. LOD usually decreases when replicate number increases until it remain unchanged. Exponential relations with (R²>0.95) between LOD of Listeria in milk by LFIA and electrochemical method with replicate number have been obtained. Finally, the review discusses the challenges and future perspectives (including the role of nanomaterials/advanced materials) to improve the analytical performance for bacterial detection.

In a hyperconnected society, computer networks play a pivotal role in all activities that permeate our daily lives. In the context of smart cities, theses networks play a pivotal role in enhancing urban efficiency and sustainability. The continuous flow of packets in these networks poses security challenges, from personal data leakage to service unavailability. Anomaly-based methods are commonly used to detect network attacks. In this context, Software-Defined Networking (SDN) emerges as a solution for anomaly detection in network traffic, facilitating comprehensive, real-time analyses and dynamic adaptation to changes. This study aims to present a systematic review of feature selection techniques and evaluate the effectiveness of attribute selection with a multi-class classifier in network anomaly detection. The objective of this paper is to inspire future research and identify trade-offs among these techniques for detecting Denial of Service (DoS) attacks. Based on the findings, the analysis of the entire dataset yields superior results in terms of precision (99%), but using the output of OneR, although resulting in a slight loss of precision compared to the complete dataset, presents the highest precision among other studied techniques, indicating a trade-off between precision and processing time efficiency.

Facial expression has lots of applications in human-computer interaction. Although feature extraction and selection have been well studied, the specificity of each expression variation is not fully explored in state-of-the-art works. In this work, the problem of multiclass expression recognition is converted into triplet-wise expression recognition. For each expression triplet, a new feature optimization model based on Action Unit (AU) weighting and patch weight optimization is proposed to represent the specificity of the expression triplet. Sparse representation based approach is then proposed to detect the active AUs of testing sample for better generalization. The algorithm achieved competitive accuracies of 89.67% and 94.09% for Jaffe and CK+ databases, respectively. Better cross-database performance has also been observed.

Object detection systems are widely used in various fields such as autonomous vehicles and facial recognition. In particular, object detection using deep learning networks enables real-time processing in low-performance edge devices, and can maintain high detection rates. However, edge devices that operate far from administrators are vulnerable to various physical attacks by malicious adversaries. In this paper, we implement a function for detecting traffic signs by using three versions of You Only Look Once (YOLO) and Faster-RCNN, which can be adopted by edge devices of autonomous vehicles. Then, assuming the role of a malicious attacker, we executed adversarial patch attacks with Adv-Patch and Dpatch. As a result of experimenting with misdetection of traffic stop signs using Adv-Patch and Dpatch, we confirmed the attacks can succeed with a high probability. To defeat these attacks, we propose an image reconstruction method using an autoencoder and the Structural Similarity Index Measure (SSIM). We confirm that the proposed method can sufficiently defend against an attack, attaining an AP above 91.46% even when two adversarial attacks are launched.

The existence of low-cost and accurate tools that can equip the First Responders (FRs) to detect trapped victims and provide insights about their health condition in the aftermath of a disaster, is imperative. To address the problem of detecting trapped victims behind walls or debris, we have developed a tool that exploits an Ultra-Wide Band (UWB) radar sensor for data collection and Machine Learning (ML) algorithms for data analysis. To evaluate the efficacy of our approach, we collected data from nine humans, both in standing and lying down positions. Next, we applied various ML algorithms to the collected dataset for two discrete sub-tasks that are of interest from an FR’s perspective. The first task is the detection of the victim’s presence, where the algorithms attained more than 95% accuracy and F1-Score. The second task is the estimation of the distance between the radar sensor and the victim, where the tool showed an average error of less than 40 cm.

More and more works are using machine learning techniques while adopting supervisory control and data acquisition (SCADA) system for wind turbine anomaly or failure detection. While parameter selection is important for modelling a wind turbine’s health condition, only a few papers have been published focusing on this issue and in those papers interconnections among sub-components in a wind turbine are used to address this problem. However, merely the interconnections for decision making sometimes is too general to provide a parameter list considering the differences of each SCADA dataset. In this paper, a method is proposed to provide more detailed suggestions on parameter selection based on mutual information. Moreover, after proving that Copula, a multivariate probability distribution for which the marginal probability distribution of each variable is uniform is capable of simplifying the estimation of mutual information, an empirical copula based mutual information estimation method (ECMI) is introduced for an application. After that, a real SCADA dataset is adopted to test the method, and the results show the effectiveness of the ECMI in providing parameter selection suggestions when physical knowledge is not accurate enough.

Isothermal Nucleic Acid Amplification based-Lateral Flow Testing (INAA-LFT) has emerged as a robust technique for on-site pathogen detection, providing a visible indication of pathogen nucleic acid amplification that rivals or even surpasses the sensitivity of real-time quantitative PCR. The isothermal nature of INAA-LFT ensures consistent conditions for nucleic acid amplification, establishing it as a crucial technology for rapid on-site pathogen detection. However, despite its considerable promise, the widespread application of isothermal INAA amplification-based lateral flow testing faces several challenges. This review provides an overview of the INAA-LFT procedure, highlighting its advancements in detecting plant viruses. Moreover, the review underscores the imperative to address existing limitations and emphasizes ongoing research efforts dedicated to enhancing the applicability and performance of this technology in the realm of rapid on-site testing.

The video images captured at long range usually have low contrast floating objects of interest on a sea surface. A comparative experimental study of the statistical characteristics of reflections from floating objects and from the agitated sea surface showed the difference in the correlation and spectral characteristics of these reflections. The functioning of the recently proposed modified matched subspace detector (MMSD) is based on the separation of the observed data spectrum on two subspaces: relatively low and relatively high frequencies. In the literature the MMSD performance has been evaluated in generally and moreover using only a sea model (additive Gaussian background clutter). This paper extends the performance evaluating methodology for low contrast object detection and moreover using only the real sea dataset. This methodology assumes an object of low contrast if the mean and variance of the object and the surrounding background are the same. The paper assumes that the energy spectrum of the object and the sea are different. The paper investigates a scenario in which an artificially created model of a floating object with specified statistical parameters is placed on the surface of a real sea image. The paper compares the efficiency of the classical Matched Subspace Detector (MSD) and MMSD for detecting low-contrast objects on the sea surface. The article analyzes the dependence of the detection probability at a fixed false alarm probability on the difference between the statistical means and variances of a floating object and the surrounding sea.

Human Face and facial parts are the most significant parts as it reveals a person’s true identity. It plays an important role in various biometric applications like crowd analysis, human tracking, photography, cosmetic surgery, etc. There are many techniques are available to detect a facial image. Among them, skin detection is the most popular one. The aim of this paper is to detect first the person's identity from facial image and finally check any spot present the the detected person. The first step is to detect the maximum skin region based on a combination method of RGB and HSV color space model. Next it is to verify the skin areas of human through machine learning approach. The Aggregated Channel Features (ACF) detector is used to identify the different facial parts like eye pairs, nose, and mouth. Bootstrap aggregation decision tree classifier is applied to classify the person’s identity based on Histogram Oriented Gradient (HOG) features value. The experimental results show that the proposed method gives the average 97% accuracy.

After long years of development, the technology of analyzing the working condition of power units based on vibration signals has had relatively stable applications, but the accuracy and the degree of automation and intelligence for fault diagnosis are still inadequate due to the limitations of the current development of key technologies. With the development of big data and artificial intelligence technology, the involvement of new technologies will be an important boost to the development of this field. To support the subsequent research, bibliometrics is used as a tool to sort out the development of the technology in this field at the macro level; at the micro level, the classical and key literature is studied to grasp the development status at the technical level and prepare for the selection of entry points to continue in-depth innovation afterwards.

Technological developments within the maritime sector are resulting in rapid progress that will see the commercial use of autonomous vessels, known as Maritime Autonomous Surface Ships (MASS). Such ships are equipped with a range of advanced technologies, such as IoT devices, artificial intelligence (AI) systems, machine learning (ML) based algorithms and augmented reality (AR) tools. Through such technologies, the autonomous vessels can be remotely controlled from Shore Control Centres (SCC) by using real-time data to optimise their operations, enhance safety and reduce the possibility of human error. Apart from the regulatory aspects, which are under definition by the International Maritime Organisation (IMO), cybersecurity aspects must be considered and properly addressed to prevent such complex systems from being tampered with. In this paper, we propose an approach to secure communication channels between the SCCs and the vessels using SECOM standard and the underlying digital infrastructure in charge of data collection, storage and processing by relying on a set of machine learning (ML) algorithms for anomaly detection. The proposed approach is validated against a real implementation of the SCC deployed at Livorno seaport premises. Finally, the experimental results and the performance evaluation are provided to assess its effectiveness accordingly.

Focusing on the problem of identifying and classifying aero-engine models, this paper measures the infrared spectrum data of aero-engine hot jets using a telemetry Fourier transform infrared spectrometer. Simultaneously, infrared spectral data sets with the six different types of aero-engines are created. For the purpose of classifying and identifying infrared spectral data, a CNN architecture based on the continuous wavelet transform peak seeking attention mechanism (CWT-AM-CNN) is suggested. This method calculates the peak value of middle wave band by continuous wavelet transform, and the peak data is extracted by the statistics of the wave number locations with high frequency. Attention mechanism is used for the peak data, and the attention mechanism is weighted to the feature map of the feature extraction block. The training set, vali-dation set and prediction set are divided in the ratio of 8:1:1 for the infrared spectral data sets. For three different data sets, CWT-AM-CNN proposed in this paper is compared with the classical classifier algorithm based on CO2 feature vector and the popular AE, RNN and LSTM spectral processing networks. The prediction accuracy of the proposed algorithm in the three data sets is as high as 97%, and the lightweight network structure design not only guarantees high precision, but also has a fast running speed, which can realize the rapid and high-precision classification of the infrared spectral data of the aero-engine hot jets.

The term "code smell" refers to an indication of a problem with the quality of source code. Numerous studies have been conducted to identify problematic features in source code. Initially, the focus was on utilizing metric-based and heuristic-based approaches. In recent years, however, there has been a shift towards using machine learning and deep learning (DL) techniques for smell detection. Nevertheless, the current algorithms are still considered to be in the early stages of development. Recognizing the challenges associated with identifying smells using DL methods, both academics and software developers have made efforts to address these obstacles. This work involves constructing and evaluating new DL models for code smell detection. Two models are built upon stacked autoencoders, employing a hybrid architecture that combines bidirectional long short-term memory and convolutional neural network components.

Abstract Plant diseases are responsible for over 10% loss in global food production and have serious impacts on world-wide food security across several countries. Amongst these diseases, those caused by viroids are highly significant, accounting for extensive damage to crops every year and pose a major economic threat to agricultural crops. Combined infections with viroids, plant viruses and their satellites further exacerbate the situation. Vector transmission of viroids necessitates the use of pesticides, thus incurring high costs in addition to eliciting several detrimental effects on the ecosystem due to the noxious outcomes of pesticide use. The current review addresses the major economic impacts of viroids on crops, costs to the world economy and impacts on regions, farmers and consumers in addition to describing remediation measures to mitigate the losses caused by these subviral agents. Also described are issues relating to the pros and cons of pesticide use, the compelling need towards sustainable agricultrual practices while reducing pesticide use. In particular, the role of RNAi and genome editing technologies towards amelioration of agricultural loss is discussed.

Deep learning has increasingly been employed to identify minerals. However, deep learning can only be used to identify minerals in the distribution of the training set, while any mineral outside the spectrum of the training set is inevitably categorized erroneously within a predetermined class from the training set. To solve this problem, the study introduces the approach that amalgamates One-Class Support Vector Machines (OCSVM) with the ResNet architecture for the out-of-distribution mineral detection. Initially, ResNet undergoes training using a training set comprising well-defined minerals. Subsequently, the earlier layers of the trained ResNet are employed to extract the discriminative features of the mineral under consideration. These extracted mineral features then become the input for OCSVM. When OCSVM discerns the mineral in the training set's distribution, it triggers the subsequent layers within the trained ResNet, facilitating the accurate classification of the mineral into one of the predefined categories encompassing the known minerals. In the event OCSVM identifies the mineral out of the training set's distribution, it is unequivocally categorized as an unclassified or 'unknown' mineral. Empirical results substantiate the method's capability to identify out-of-distribution minerals while concurrently maintaining a commendably high accuracy rate for the classification of the 36 in-distribution minerals.

The existing segmentation-based scene text detection methods mostly need complicated post-processing, and the post-processing operation is separated from the training process, which greatly reduces the detection performance. The previous method, DBNet successfully simplified post-processing and integrated the post-processing into a segmentation network. However, the training process of the model took a long time for 1200 epochs and the sensitivity to texts of various scales was lacking, leading to some text instances being missed. Considering the above two problems, we design the text detection Network with Binarization of Hyperbolic Tangent(HTBNet). First of all, we propose Binarization of Hyperbolic Tangent (HTB), optimized along with which, the segmentation network can expedite the initial convergent speed by reducing the amount of epochs from 1200 to 600. Because features of different channels in the same scale feature map focus on the information of different regions in the image, to better represent the important features of all objects in the image, we devise the Multi-Scale Channel Attention(MSCA). Meanwhile considering that multi-scale objects in the image cannot be simultaneously detected, we propose a novel module named Fused Module with Channel and Spatial(FMCS), which can fuse the multi-scale feature maps from channel and spatial dimension. Finally we adopt cross entropy as the loss function, which measures the difference between predicted values and ground truths. The experimental results show that HTBNet compared with lightweight models has achieved competitive performance and speed on Total-Text(F-measure:86.0%, FPS:30) and MSRA-TD500 (F-measure:87.5%, FPS:30).

In this study, we explored the application of deep learning techniques for credit card fraud detection, aiming to improve the performance and reliability of anomaly detection methods in financial transactions. We first utilized the Isolation Forest algorithm, achieving a detection accuracy of 26% for the top 1000 transactions. Subsequently, we experimented with the Autoencoder algorithm, an unsupervised deep neural network model, which enhanced the detection accuracy to 33.6% in the best case despite some fluctuations. The results demonstrate deep learning models' strong feature extraction capability and adaptability, highlighting their potential to surpass traditional methods. However, the high imbalance in the dataset, with only 0.17% of transactions being fraudulent, poses a significant challenge. This study underscores the necessity for further experimentation and optimization of network structures and hyperparameters to achieve more stable and efficient fraud detection. The findings provide valuable insights and reference points for future research in financial fraud detection using deep learning methodologies.

The Doppler band compensated by the keystone transform (KT) is limited. Therefore, it needs to be used in conjunction with the Doppler ambiguity compensation function to correct the range migration (RM) caused by maneuvering targets with Doppler ambiguity. However, the KT implemented by sinc interpolation suffers from significant performance loss at boundaries of compensation Doppler bands. Additionally, in a multi-target scenario, KT implementation methods occupy high complexity when the Doppler range of targets spans over two compensation Doppler bands. To address aforementioned issues, this study presents a variable Doppler starting point keystone transform (VDSPKT) method, where a new form of ambiguity compensation function is constructed, turning the Doppler starting point of the compensation band in KT variable. Firstly, the position of the compensation Doppler band is changed from fixed to adjustable as needed, enhancing the flexibility of KT. Crucially, the connection points of the compensation Doppler bands in sinc interpolation are reset, avoiding the performance loss at their boundaries. Also, the compensation band is adjusted to cover the narrow Doppler frequency range caused by targets, significantly improving the computational efficiency. Finally, the simulation results demonstrate that the proposed approach effectively addresses the performance degradation and high computational complexity of KT in the aforementioned scenarios, resulting in a computational load reduced by approximately 50% compared to traditional methods.

Semidiurnal tidal currents can exceed 5 ms−1 in Minas Passage, Bay of Fundy, where a tidal energy demonstration area has been designated to generate electricity using marine hydrokinetic turbines. The risk of harmful fish-turbine interaction cannot be dismissed for either migratory or local fish populations. Individuals belonging to several fish populations have been acoustically-tagged and monitored by using acoustic receivers moored within Minas Passage. Detection efficiency ρ is required as the first step to estimate probability of fish-turbine encounter. Moored Innovasea HR2 receivers and high residency (HR) tags were used to obtain detection efficiency ρ as a function of range and current speed, for near-seafloor signal paths within the tidal energy development area. Strong tidal currents moved moorings so HR tag signals, and their reflections from the sea surface, were used to measure ranges from tags to receivers. HR2 self-signals that reflected off the sea surface showed which moorings were displaced to lower and higher levels on the seafloor. Some of the range testing paths had anomalously low ρ which might be attributed to variable bathymetry blocking the line of sight signal path. Clear and blocked signal paths accord with mooring levels. Application of ρ is demonstrated for calculation of abundance, effective detection range, and detection-positive intervals. High residency signals were better detected than pulse position modulation (PPM) signals. Providing the presently obtained ρ applies to tagged fish that swim higher in the water column, there is a reasonable prospect that probability of fish-turbine encounter can be estimated by monitoring fish that carry HR tags.

The severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2), etiological agent of the novel coronavirus disease 2019 (COVID-19), has spread since December 2019, resulting in massive health and economic crisis worldwide. While efforts to stop the pandemic are crucial, collecting epidemiological data to help manage current and future pandemics will be important. In addition to humans, serological and molecular based studies have demonstrated SARS CoV-2 exposure in several wild, domestic and farmed animals. For examples Shriner and the team showed serologically an exposure of 40% to the white deer living in close proximity to urban centers. Additional reports have also emerged of susceptibility of animal’s species like cats, ferrets, raccoon dogs, cynomolgus macaques, rhesus macaques, white-tailed deer, rabbits, Egyptian fruit bats, and Syrian hamsters to SARS-CoV-2 infection.. It’s worth emphasizing that these reports are based on experimental data mostly derived from Europe, USA, South America and parts of Asia. In limited instances natural infections of SARS-CoV-2 have been reported in pet dogs, cats, tigers, lions, snow leopards, pumas, gorillas at zoos and farmed mink and ferrets. The presence of the virus in animal species and an understanding of whether these are natural or recent human to animal transmissions is important. It’s possible that such transmission could passage the virus or subject the virus to a different immunological pressure thereby helping with the development of viral variants in addition to being a host for future reservoirs of the virus. In Kenya SARS-CoV-2 was first detected on March 12th 2020 from imported human cases of persons who had travelled from the United States. This was followed by detection of imported cases majorly from China, Sweden and United Kingdom. Later infections were confirmed in Nairobi and Mombasa suggesting further cases of disease importations through the major ports of entry. However, no comparable data on animal exposure have hitherto been generated in Kenya. To address this key concern, we focused on three objectives; 1) development of a robust antibody ELISA based on crude SARS-CoV-2 lysate. 2) SARS-CoV-2 serology of domestic animals in Kenya. 3) Corroboration of the crude lysate based seroprevalence data and a commercial ELISA kit based on the Spike receptor binding domain (RBD) antigen. Our sample set included camel sera (both pre- & post outbreak sera), as well as sera from cats and dogs collected at the peak of the pandemic. Our results using the ELISA based on crude SARS-CoV-2 lysate indicated SARS-CoV-2 antibodies in camels (71%, N=145), cats 11% (N=16) and dogs (81%, N=36) with varying titer levels. These findings were comparable to those obtained using the commercial ELISA kit based on the spike RBD antigens. In summary, the data warrants two key conclusions: (i) we have demonstrated that the crude lysate ELISA allows for SARS-CoV-2 antibody detection, and given its potential to offer robust detection could be applied for initial mass screening (ii) although the current study cannot disentangle the relative contributions of antigenic cross-reactivity, pre-pandemic exposure to SARS-CoV-2 or human-animal transmission, it nonetheless demonstrates for the first time the prevalence of SARS-CoV-2 like antibodies in domestic and wild animals in Kenya. Our findings set the scene for further research into the prevalence of SARS-CoV-2 in domestic and wild animals to understand their potential epidemiological implications.

YOLOv8, as an efficient object detection method, can swiftly and precisely identify objects within images. However, traditional algorithms encounter difficulties when detecting small targets in remote sensing images, such as missing information, background noise, and interactions among multiple objects in complex scenes, which may affect performance. To tackle these challenges, we propose an enhanced algorithm optimized for detecting small objects in remote sensing images, named HP-YOLOv8. Firstly, we design the C2f-D-Mixer (C2f-DM) module as a replacement for the original C2f module. This module integrates both local and global information, significantly improving the ability to detect features of small objects. Secondly, we introduce a feature fusion technique based on attention mechanisms, named Bi-Level Routing Attention in Gated Feature Pyramid Network (BGFPN). This technique utilizes an efficient feature aggregation network and reparameterization technology to optimize information interaction between different scale feature maps, and through the Bi-level Routing Attention (BRA) mechanism, it effectively captures critical feature information of small target objects. Finally, we propose the Smooth Mean Perpendicular Distance Intersection over Union (SMPDIoU) loss function. The method comprehensively considers the shape and size of detection boxes, enhances the model's focus on the attributes of detection boxes and provides a more accurate bounding box regression loss calculation method. To demonstrate our approach's efficacy, we conduct comprehensive experiments across the RSOD, NWPU VHR-10, and VisDrone2019 datasets. The experimental results show that the HP-YOLOv8 achieves 95.11%, 93.05%, and 53.49% in the [email protected] metric, and 72.03%, 65.37%, and 38.91% in the more stringent [email protected]:0.95 metric, respectively.

Using coherent optical detection and digital signal processing, laser phase noise and equalization enhanced phase noise can be effectively mitigated using the feed-forward and feed-back carrier phase recovery approaches. In this paper, theoretical analyses of feed-back and feed-forward carrier phase recovery methods have been carried out in the long-haul high-speed n-level phase shift keying (n-PSK) optical fiber communication systems, involving a one-tap normalized least-mean-square (LMS) algorithm, a block-wise average algorithm, and a Viterbi-Viterbi algorithm. The analytical expressions for evaluating the estimated carrier phase and for predicting the bit-error-rate (BER) performance (such as the BER floors) have been presented and discussed in the n-PSK coherent optical transmission systems by considering both the laser phase noise and the equalization enhanced phase noise. The results indicate that the Viterbi-Viterbi carrier phase recovery algorithm outperforms the one-tap normalized LMS and the block-wise average algorithms for small phase noise variance (or effective phase noise variance), while the one-tap normalized LMS algorithm shows a better performance than the other two algorithms for large phase noise variance (or effective phase noise variance). In addition, the one-tap normalized LMS algorithm is more sensitive to the level of modulation formats.

In the current paper, a residual generation filter with finite memory structure is proposed for sensor fault detection. The proposed finite memory residual generation filter provides the residual by real-time filtering of fault vector using only the most recent finite observations and inputs on the window. It is shown that the residual given by the proposed residual generation filter provides the exact fault for noise-free systems. The proposed residual generation filter is specified to the digital filter structure for the amenability to hardware implementation. Finally, to illustrate the capability of the proposed residual generation filter, numerical examples are performed for the discretized DC motor system having the multiple sensor faults.

pH value almost affects the function of cells and organisms in all aspects, so in biology, biochemical and many other research fields, it is necessary to apply simple, intuitive, sensitive, stable detection of pH and base characteristics inside and outside the cell. Therefore, many research groups have explored the design and application of pH probes based on surface enhanced Raman scattering（SERS）. In this review article, we discussed the basic theoretical background of explaining the working mechanism of pH SERS sensors, and also briefly described the significance of cell pH measurement, and simply classified and summarized the factors that affected the performance of pH SERS probes. Some applications of pH probes based on surface enhanced Raman scattering (SERS) in intracellular and extracellular pH imaging and the combination of other analytical detection techniques are described. And finally, the development prospect of this field is prospected.

To secure computers and information systems from attackers taking advantage of vulnerabilities in the system to commit cybercrime, several methods have been proposed for real-time detection of vulnerabilities to improve security around information systems. Of all the proposed methods, machine learning had been the most effective method in securing a system with capabilities ranging from early detection of software vulnerabilities to real-time detection of ongoing compromise in a system. As there are different types of cyberattacks, each of the existing state-of-the-art machine learning models depends on different algorithms for training which also impact their suitability for detection of a particular type of cyberattack. In this research, we analyzed each of the current state-of-theart machine learning models for different types of cyberattack detection from the past 10 years with a major emphasis on the most recent works for comparative study to identify the knowledge gap where work is still needed to be done with regard to detection of each category of cyberattack

Hyperspectral small target detection (HSTD) is a promising pixel-level detection task. However, due to the low contrast and imbalance number between the target and background spatially and high dimensions spectrally, it is a challenging one. To address these issues, this work proposes a representation learning-based graph and generative network for hyperspectral small target detection. The model builds a fusion network through frequency representation for HSTD, where the novel architecture incorporates irregular topological data and the spatial-spectral feature to improve its representation ability. Firstly, a graph convolutional network (GCN) module better models the non-local topological relationship between samples to represent the hyperspectral scene’s underlying data structure. The mini-batch-training pattern of the GCN decreases the high computational cost of building an adjacency matrix for high-dimensional data sets. In parallel, the generative model enhances the differentiation reconstruction and the deep feature representation ability with respect to the target spectral signature. Finally, a fusion module compensates for the extracted different types of HS features and integrates their complementary merits for hyperspectral data interpretation while increasing detection and background suppression capabilities. Experiments on different hyperspectral data sets demonstrate the advantages of the proposed architecture.

Quasi-bound state in the continuum (QBIC) can effectively enhance the interaction of terahertz (THz) wave with matters due to the tunable high-Q property, which has a strong potential application in detection of low-concentration biological samples in the THz band. In this paper, a novel THz metamaterial sensor with a double-chain separated resonant cavity structure based on QBIC is designed and fabricated. The process of excitation of the QBIC mode is verified and the structural parameters are optimized after considering the ohmic loss by simulations. The simulated refractive index sensitivity of the sensor is up to 544 GHz/RIU, much higher than those of recently reported THz metamaterial sensors. Sensing sensitivity of the proposed metamaterial sensor is confirmed in experiment by detecting low-concentration lithium citrate (LC) and bovine serum albumin (BSA) solutions. The limits of detection (LoDs) are obtained to be 0.0025 mg/mL (12 μM) for LC and 0.03125 mg/mL (0.47 μM) for BSA, respectively, both of which excel over most of reported results in previous studies. These results indicate that the proposed THz metamaterial sensor has excellent sensing performances and can well be applied to the detection of low concentration biological samples.

We used an ultrasensitive, broadband optomechanical ultrasound sensor to study the acoustic signals produced by pressurized nitrogen escaping from a variety of small syringes. Harmonically related jet tones extending into the MHz region were observed for a certain range of flow (i.e., Reynolds number), in qualitative agreement with historical studies on gas jets emitted from pipes/orifices of much larger dimensions. For higher turbulent flow rates, we observed broadband ultrasonic emission in the ~ 0 – 5 MHz range, likely limited on the upper end by attenuation in air. These observations are made possible by the broadband, ultrasensitive response (for air-coupled ultrasound) of our optomechanical devices. Aside from their theoretical interest, our results could have practical implications for the non-contact monitoring and detection of early-stage leaks in pressured fluid systems.

Ship detection from synthetic aperture radar (SAR) imagery is crucial for various fields in real-world applications. Numerous deep learning-based detectors have been investigated for SAR ship detection, which require a substantial amount of labeled data for training. However, SAR data annotation is time-consuming and demands specialized expertise, resulting in that deep learning-based SAR ship detectors struggle due to a lack of annotations. With limited labeled data, semi-supervised learning is a popular approach for boosting detection performance by excavating valuable information from unlabeled data. In this paper, a semi-supervised SAR ship detection network is proposed, termed as Multi-Teacher Dempster-Shafer Evidence Fusion Net-work (MTDSEFN). The MTDSEFN is an enhanced framework based on the basic teacher-student skeleton frame, comprising two branches: the Teacher Group (TG) and the Agency Teacher (AT). The TG utilizes multiple teachers to generate pseudo-labels for different augmentation versions of unlabeled samples, which are then refined to obtain high-quality pseudo-labels by using Dempster-Shafer (D-S) fusion. The AT not only serves to deliver weights of its own teacher to the TG at the end of each epoch, but also updates its own weights after each iteration, enabling the model to effectively learn rich information from unlabeled data. The combination of TG and AT guarantees both reliable pseudo-label generation and learning comprehensive diversity information from numerous unlabeled samples. Extensive experiments were performed on two public SAR ship datasets, and the results demonstrated the effectiveness and superiority of the proposed approach.

Food safety is a significant issue that affects people worldwide and is tied to their lives and health. The issue of pesticide residues in food is just one of many issues related to food safety, which leave residues in crops and are transferred through the food chain to human consumption. Foods contaminated with pesticide residues pose a serious risk to human health, including carcinogenicity, neurotoxicity and endocrine disruption. Although traditional methods including gas chromatography, high-performance liquid chromatography, chromatography and mass spectrometry can be used to achieve quantitative analysis of pesticide residues, the disadvantages of these techniques, such as being time-consuming and costly and requiring specialist staff, limit their application. Therefore, there is a need to develop rapid, effective and sensitive equipment for the quantitative analysis of pesticide residues in food. Microfluidics is rapidly emerging in a number of fields due to its outstanding strengths. This paper summarizes the application of microfluidic techniques to pyrethroid, carbamate, organochlorine and organophosphorus pesticides, as well as to commercial products. Meanwhile, the paper also outlines the development of microfluidics in combination with 3D printing technology and nanomaterials for the detection of pesticide residues in food.

Texture is essential information for image representation, capturing patterns, and structures. Consequently, texture plays a crucial role in the manufacturing industry and has been extensively studied in the fields of computer vision and pattern recognition. However, real-world textures are susceptible to defects, which can degrade the image quality and cause various issues. Therefore, there is a need for accurate and effective methods to detect texture defects. In this study, a simple autoencoder and Fourier transform were employed for texture defect detection. The proposed method combines Fourier transform analysis with the reconstructed template obtained from the simple autoencoder. Fourier transform is a powerful tool for analyzing the frequency domain of images and signals. Moreover, analyzing the frequency domain enables effective defect detection because texture defects often exhibit characteristic changes in specific frequency ranges. The proposed method demonstrates effectiveness and accuracy in detecting texture defects. Experimental results are presented to evaluate its performance and compare it with those of existing approaches.

Identifying punch types as part of punch kinematic analysis is a key component for both coaches and players, providing critical insights into the variety and effectiveness of punches, which are essential for refining overall performance. Existing research indicates that punch type classification typically relies on data from either wearables or video data. Both approaches have their merits and demerits but require a significant amount of labeled data for modeling the classification algorithm. In this work, we propose a novel approach that significantly reduces the labeling effort by one-sixth by leveraging video data to enhance the training information of sensor data. This enriched training phase enables more accurate punch classification with reduced labeled data. Subsequently, the real-time punch identification and classification use only the wearable data. On a real dataset, with 15% of the labeling efforts, our approach achieved an accuracy rate of 91.41% for rear hand punch recognition and 91.91% for lead hand punch recognition, along with 92.33% and 94.56% for punch classification, respectively. Through the development of our Smart Boxer system, we aim to revolutionize punch analytics in boxing, providing comprehensive insights while enhancing the overall training and competitive experience for athletes and enriching fan engagement with the sport.

Synthetic Aperture Radar (SAR) ship target detection has been extensively researched. However, most methods use the same dataset division for both training and validation. In practical applications, it is often necessary to quickly adapt to new loads, new modes, and new data to detect targets effectively. This presents a cross-domain detection problem that requires further study. This paper proposes a method for detecting SAR ships in complex backgrounds using fusion tensor and cross-domain adversarial learning. The method is designed to address the cross-domain detection problem of SAR ships with large differences between the training and test sets. Specifically, it can be used for the cross-domain detection task from the fully-polarised medium-resolution ship dataset (source domain) to the high-resolution single-polarised dataset (target domain). This method proposes a Channel Fusion Module (CFM) based on the YOLOV5s model. The CFM utilises the correlation between polarised channel images during training to enrich the feature information of single-polarised images extracted by the model during inference. The article proposes a module called Cross-Domain Adversarial Learning Module (CALM) to reduce overfitting and achieve adaptation between domains. Additionally, the paper introduces the Anti-Interference Head (AIH) which decouples the detection head to reduce the conflict of classification and localization problems. This improves the anti-interference and generalization ability in complex backgrounds. This paper conducts cross-domain experiments using the constructed medium-resolution SAR full polarization dataset (SFPD) as the source domain and the high-resolution single-polarised ship detection dataset (HRSID) as the target domain. Compared to the best-performing YOLOV8s model among typical mainstream models, this model improves Precision by 4.9%, Recall by 3.3%, AP by 2.4%, and F1 by 3.9%. This verifies the effectiveness of the method and provides a useful reference for improving cross-domain learning and model generalization capability in the field of target detection.

A recent study showed the potential of DA Perten 7200 NIR Spectrometer in detecting chlorpyrifos-methyl pesticide residue in rough, brown, and milled rice. However, this instrument is still lab-based and generally suited for point-of-sale testing in many countries. To provide a field deployable version of this technique, an existing light emitting diode (LED)-based instrument which provide discrete NIR wavelength illumination and reflectance spectra over the range of 850-1550 nm was tested. Spectra were collected from rough, brown, and milled rice at different pesticide concentrations and analysed for quantitative and qualitative measurement using partial least squares regression (PLS) and discriminant analysis (DA). Simulations for LED-based instruments were also evaluated using segments of spectra from the DA7200 to represent LED illumination. For the simulation of the existing LED-based instrument (LEDPrototype1) using their wavelengths range yielded 70.4% to 100% correct classification. Simulation of a second LED instrument, LEDPrototype2, with spectral segments selected based on significant wavelength regions from PLS regressions coefficients obtained from the DA7200 showed improved performance with R2 of 0.59 to 0.82 and correct classifications of 71.3 to 100%. An actual LED based instrument with this capability could provide a quick screening tool to determine if MRLs are exceeded.

Existing methods of hyperspectral anomaly detection still face several challenges: 1) Due to the limitations of self-supervision, avoiding the identity mapping of anomalies remains difficult; 2) The ineffective interaction between spatial and spectral features leads to the insufficient utilization of spatial information; 3) Current methods are not adaptable to the detection of multi-scale anomaly targets. To address the aforementioned challenges, we proposed a blind spot network based on multi-scale blind spot convolution for HAD. The multi-scale mask convolution module is employed to adapt to diverse scales of anomaly targets, while the dynamic fusion module is introduced to integrate the advantages of mask convolutions at different scales. The proposed approach includes a spatial-spectral joint module and a background feature attention mechanism to enhance the interaction between spatial-spectral features, with a specific emphasis on highlighting the significance of background features within the network. Furthermore, we propose a preprocessing technique that integrates pixel shuffle down-sampling (PD) operation with spatial spectral joint screening to address the issue of anomalous identity map-ping and facilitate finite-scale mask convolution to adapt to the detection of more scale targets. The proposed approach was assessed on four real hyperspectral datasets comprising anomaly targets of different scales. The experimental results demonstrate the effectiveness and superior performance of the proposed methodology compared with nine state-of-the art methods.

The use of strobilurin fungicides in agriculture has increased steadily during the past 25 years, and although strobilurins have minimal water solubility, they regularly appear in surface waters, at times in concentrations approaching toxic levels for aquatic life. The present study examined concentrations of strobilurin fungicides in designated trout streams draining an agricultural watershed in southeastern Minnesota, USA, where fungicides may have contributed to a recent fish kill. Water samples (n = 131) were analyzed for the presence of five different strobilurin fungicides (azoxystrobin, fluoxastrobin, picoxystrobin, pyraclostrobin, trifloxystrobin). Samples were collected via grab and automated sampling during baseflow and stormflow events throughout an entire crop growing season from sites on each of three forks of the Whitewater River. Detection frequencies for the five strobilurins ranged from 44 to 82%. Fluoxastrobin and pyraclostrobin concentrations were above known toxic levels in 3% and 15% of total samples analyzed, respectively. The highest concentrations were detected in mid-summer (mid-June to mid-August) samples, coincident with likely strobilurin applications. Lower concentrations were present in water samples collected during the non-application periods in spring and fall, suggesting groundwater-stream interactions or steady leaching of fungicides from watershed soils or stream sediments. Further study is required to determine strobilurin concentrations in sediments, soils, and groundwater. Better tracking and guidance regarding strobilurin use is necessary to adequately protect aquatic life as fungicide use continues to increase.

In the field of autonomous vehicles, lane detection and control plays an important role. In autonomous driving the vehicle has to follow the path to avoid the collision. A deep learning technique is used to detect the curved path in autonomous vehicles. In this paper a customized lane detection algorithm was implemented to detect the curvature of the lane. A ground truth labelling tool box for deep learning is used to detect the curved path in autonomous vehicle. By mapping point to point in each frame 80-90% computing efficiency and accuracy is achieved in detecting path.

Recent years have witnessed a rapid growth of code-reuse attacks in advance persistent threats and cyberspace crimes. Carefully crafted code-reuse exploits circumvent modern protection mechanisms and hijack the execution flow of a program to perform expected functionalities by chaining together existing codes. The sophistication and intrication of code-reuse exploits hinder the scrutinization and dissection on them. Although the previous literature has introduced some feasible approaches, effectiveness and reliability in practical applications remain severe challenges. To address this issue, we propose Horus, a data-driven framework for effective and reliable detection on code-reuse exploits. In order to raise the effectiveness against underlying noises, we comprehensively leverage the strengths of time-series and frequency-domain analysis, and propose a learning-based detector that synthesizes the contemporary twofold features. Then we employ a lightweight interpreter to speculatively and tentatively translate the suspicious bytes to open the black box and enhance the reliability and interpretability. Additionally, a functionality-preserving data augmentation is adopted to increase the diversity of limited training data and raise the generality for real-world deployment. Comparative experiments and ablation studies are conducted on a dataset composed of real-world instances to verify and prove the prevalence of Horus. The experimental results illustrate that Horus outperform existing methods on the identification of code-reuse exploits from data stream with an acceptable overhead. Horus does not rely on any dynamic executions and can be easily integrated into existing defense systems. Moreover, Horus is able to provide tentative interpretations about attack semantics irrespective of target program, which further improve system's effectiveness and reliability.

The development of Artificial Intelligence-based tools is having a big impact on industry. In this context, the maintenance operations of important assets and industrial resources are deeply changing both from a theoretical and a practical perspective, even in the field of ship propulsion systems. Namely, conventional preventive maintenance schedules audits and checks at fixed times by using statistics on component failures, but it can be improved by a predictive maintenance based on the real component health status, which is inspected by proper sensors. In this way, time and costs are saved. More in details, data-driven models, through Machine Learning (ML) algorithms, generate the expected values of monitored variables for comparison with real measurements coming from the asset, hence for a diagnosis based on the difference between expectations and observations. Then, a ML-based fault detection starts with the choice of the ML algorithm. This process is often not the consequence of an in-depth deep analysis of the different algorithms available in the literature. For that reason, the authors propose a simple procedure to support a first implementation stage of the so-called Condition Based Maintenance (CBM): a quick benchmarking of the most suitable ML algorithms useful for fault detection. The algorithms are compared by taking into account not only the algorithm output error, shown by the Mean Squared Error between predictions and data, but also the response time, which is a crucial variable for maintenance purpose.

An existing power grid is going through a massive transformation. Smart grid technology is a radical approach for improvisation in prevailing power grid. Integration of electrical and communication infrastructure is inevitable for the deployment of Smart grid network. Smart grid technology is characterized by full duplex communication, automatic metering infrastructure, renewable energy integration, distribution automation and complete monitoring and control of entire power grid. Wireless sensor networks (WSNs) are small micro electrical mechanical systems which are accomplished to collect and communicate the data from surroundings. WSNs can be used for monitoring and control of smart grid assets. Security of wireless sensor based communication network is a major concern for researchers and developers. The limited processing capabilities of wireless sensor networks make them more vulnerable to cyber-attacks. The countermeasures against cyber-attacks must be less complex with an ability to offer confidentiality, data readiness and integrity. The address oriented design and development approach for usual communication network requires a paradigm shift to design data oriented WSN architecture. WSN security is an inevitable part of smart grid cyber security. This paper is expected to serve as a comprehensive assessment and analysis of communication standards, cyber security issues and solutions for WSN based smart grid infrastructure.

Camouflage evaluation typically involves human visual search and detection experiments that are time-consuming and expensive. Hence, there is a need for models that compute camouflage effectiveness from digital imagery. Convolutional neural networks are a powerful tool for automatic object detection and recognition. We investigated whether such a network (YOLO) can also provide a measure of camouflage effectiveness that is related to human perception. To this end, human performance measures of camouflage effectiveness such as detection time and target conspicuity were obtained in observer experiments and compared with the performance of the (standard, i.e. pre-trained) YOLO-V4-tiny algorithm. YOLO provides the probability that a detected object is correctly classified, and this is adopted as our measure of camouflage effectiveness. We compared the YOLO-predicted classification probability for a person in camouflage clothing moving through rural and urban scenes to human detection performance. Overall, camouflage effectiveness predicted by YOLO correlates with human observer performance. At close distances, YOLO’s classification performance appears sensitive to high-contrast shape-breaking elements of a camouflage pattern. This suggests that YOLO may be adapted to assess camouflage effectiveness for a wide range of applications, such as evaluating or optimizing one’s signature and predicting optimal hiding locations in a given environment. Further research is required to fully establish YOLO’s limitations and applicability for this purpose.