Sign Language Recognition (SLR) aims to bridge speech-impaired and general communities by recognizing signs from given videos. Researchers still face challenges developing efficient SLR systems because of the video’s complex background, light illumination, and subject structures. Recently many researchers developed a skeleton-based sign language recognition system to overcome the subject and background variation of hand gesture signs. However, skeleton-based SLR is still under exploration due to the lack of information and annotations on hand key points. More recently, researchers included body and face information with the hand gesture for the SLR, but their performance and efficiency are unsatisfactory. We proposed a Multi-Stream Graph-based Deep Neural Network (SL-GDN) for a skeleton-based SLR system to overcome the problems. The main purpose of the proposed SL-GDN approach is to improve the efficiency and performance of the SLR system with a low computational cost based on the human body pose in the form of 2D landmark locations. In the procedure, firstly, we constructed a skeleton graph based on the selected 27 whole-body key points among 67 key points to solve the inefficiency problems. Then we proposed multi-stream SL-GDN to extract features from the whole-body skeleton graph for four streams. Finally, we concatenated the four different features and applied a classification module to refine the feature and recognize corresponding sign classes. Our data-driven and graph construction method increases the system’s flexibility and brings high generability to adapt various data samples. We used three large-scale benchmark SLR datasets to evaluate the proposed model: WLASL, AUTSL and CSL. The demonstrated performance accuracy table proved the superiority of the proposed model, and we believe this will be considered a great invention in the SLR domain.

In recent times, with the increase of Artificial Neural Network (ANN), deep learning has brought a dramatic twist in the field of machine learning by making it more Artificial Intelligence (AI). Deep learning is used remarkably used in vast ranges of fields because of its diverse range of applications such as surveillance, health, medicine, sports, robotics, drones etc. In deep learning, Convolutional Neural Network (CNN) is at the center of spectacular advances that mixes Artificial Neural Network (ANN) and up to date deep learning strategies. It has been used broadly in pattern recognition, sentence classification, speech recognition, face recognition, text categorization, document analysis, scene, and handwritten digit recognition. The goal of this paper is to observe the variation of accuracies of CNN to classify handwritten digits using various numbers of hidden layer and epochs and to make the comparison between the accuracies. For this performance evaluation of CNN, we performed our experiment using Modified National Institute of Standards and Technology (MNIST) dataset. Further, the network is trained using stochastic gradient descent and the backpropagation algorithm.

Nowadays, the Internet of Things (IoT) concept plays a pivotal role in society and brings new capabilities to different industries. The number IoT solutions in areas such as transportation and healthcare is increasing and new services are under development. In the last decade, society has experienced a drastic increase in IoT connections. In fact, IoT connections will increase in the next few years across different areas. Conversely, despite these benefits, several challenges still need to be faced to enable efficient and secure operations (e.g., interoperability, security, standards, and server technologies). Furthermore, although efforts have been made to produce datasets composed of attacks against IoT devices, several possible attacks are not considered. Most existing efforts do not consider an extensive network topology with real IoT devices. The main goal of this research is to propose a novel and extensive IoT attack dataset to foster the development of security analytics applications in real IoT operations. To accomplish this, 33 attacks are executed in an IoT topology composed of 105 devices. These attacks are classified into seven categories, namely DDoS, DoS, Recon, Web-based, Brute Force, Spoofing, and Mirai. Finally, all attacks are executed by malicious IoT devices targeting other IoT devices.

Although various vaccines are now commercially available, they have not been able to stop the spread of COVID-19 infection completely. An excellent strategy to quickly get safe, effective, and affordable COVID-19 treatment is to repurpose drugs that are already approved for other diseases as adjuvants along with the ongoing vaccine regime. The process of developing an accurate and standardized drug repurposing dataset requires a considerable level of resources and expertise due to the commercial availability of an extensive array of drugs that could be potentially used to address the SARS-CoV-2 infection. To address this bottleneck, we created the CoviRx platform. CoviRx is a user-friendly interface that provides access to the data, which is manually curated for COVID-19 drug repurposing data. Through CoviRx, the data curated has been made open-source to help advance drug repurposing research. CoviRx also encourages users to submit their findings after thoroughly validating the data, followed by merging it by enforcing uniformity and integ-rity-preserving constraints. This article discusses the various features of CoviRx and its design principles. CoviRx has been designed so that its functionality is independent of the data it dis-plays. Thus, in the future, this platform can be extended to include any other disease X beyond COVID-19. CoviRx can be accessed at www.covirx.org.

Human Pose Estimation (HPE) is a critical technology in computer vision with diverse applications ranging from healthcare to sports analysis. This project presents a method for detecting the 2D stance of multiple persons in an image using a nonparametric representation known as Part Affinity Fields (PAFs). By leveraging the first 10 layers of the VGG-19 convolutional neural network and training on the COCO dataset, our model effectively identifies and associates key points of the human body.The architecture employs a two-branch system that jointly learns part locations and their associations through sequential prediction. This enables the model to maintain real-time performance while achieving high accuracy, regardless of the number of persons in the image. To enhance accessibility, we developed a mobile application using Flutter and TensorFlow Lite, allowing real-time pose estimation via a mobile device’s front camera. The app provides immediate feedback on physical exercises and yoga poses, making it an invaluable tool for fitness enthusiasts and healthcare professionals. Visual outputs such as heatmaps and PAFs confirm the model’s capability to accurately localize and connect key points. Despite potential challenges such as data quality and hyperparameter tuning, the results indicate that our approach is both reliable and practical for real-world deployment. This project not only advances the state-of-the-art in HPE but also opens up possibilities for future enhancements, including integrating 3D pose estimation and applying the technology in augmented and virtual reality applications.

Several studies in computer vision have examined specular removal, which is crucial for object detection and recognition. This research has traditionally been divided into two tasks: specular highlight removal, which focuses on removing specular highlights on object surfaces, and reflection removal, which deals with specular reflections occurring on glass surfaces. In reality, however, both types of specular effects often coexist, making it a fundamental challenge that has not been adequately addressed. Recognizing the necessity of integrating specular components handled in both tasks, we constructed a Specular-Light (S-Light) DB for training single-image-based deep learning models. Moreover, considering the absence of benchmark datasets for quantitative evaluation, the Multi-Scale Normalized Cross Correlation (MS-NCC) metric, which considers the correlation between specular and diffuse components, was introduced to assess the learning outcomes.

In small data set learning, obtaining a large number of training samples from the source class for transfer learning is a challenge for fine-grained visual classification. Based on the fact that fine-grained concepts can be learned with very few samples, we use very few labeled samples (for example, three) in each category. However, due to the difficulty in distinguishing the subtle differences between fine-grained images, we propose a local structure information extraction method of SDFGVC based on spatial frequency features. The proposed learning module enhances the ability of the network to find significant regions and shows excellent performance in experiments on six datasets.

As the world increasingly becomes more interconnected, the demand for safety and security is ever-increasing, particularly for industrial networks. This has prompted numerous researchers to investigate different methodologies and techniques suitable for intrusion detection systems (IDS) requirements. Over the years, many studies have proposed various solutions in this regard including signature-based and machine-learning (ML) based systems. More recently, researchers are considering deep learning (DL) based anomaly detection approaches. Most proposed works in this research field aimed to achieve either one or a combination of high accuracy, considerably low false alarm rates (FARs), high classification specificity and detection sensitivity, achieving lightweight DL models, or other ML and DL-related performance measurement metrics. In this study, we propose a novel method to convert a raw dataset to an image dataset to magnify patterns. Based on this we devise an anomaly detection for IDS using a lightweight convolutional neural network (CNN) that classifies denial of service and distributed denial of service. The proposed methods were evaluated using a modern dataset, CSE-CIC-IDS2018, and a legacy dataset, NSL-KDD. We have also applied a combined dataset to assess the generalization of the proposed model across various datasets. Our experimental results have demonstrated that the proposed methods achieved high accuracy and considerably low FARs with high specificity and sensitivity. The resulting loss and accuracy curves have also demonstrated the excellent generalization of the proposed lightweight CNN model, effectively avoiding overfitting. This holds for both the modern and legacy datasets, including their mixed version.

In this paper, intrusion detection systems are thoroughly investigated utilizing the CSE-CIC-IDS-2018 dataset. The research is divided into three key phases: first, applying Data Cleaning, Exploratory Data Analysis, and Data Normalization techniques (min-max and z-score) for preparing data across distinct classifiers. Second, feature importance is reduced using a combination of Principal Component Analysis (PCA) and Random Forest (RF), with the goal of improving processing speed and decreasing model complexity. This stage comprises a comparison with the entire dataset. Finally, machine learning algorithms (XGBoost, CART, DT, KNN, MLP, RF, LR, and Bayes) are applied to specific features and preprocessing approaches. Surprisingly, the XGBoost, DT, and RF models outperform in both ROC values and CPU runtime. Following evaluation, which includes PCA and RF feature selection, an optimal set is produced.

COVID-19 is a severe global problem, and AI can play a significant role in preventing losses by monitoring and detecting infected persons in early-stage. This paper aims to propose a high-speed and accurate fully-automated method to detect COVID-19 from the patient's CT scan images. We introduce a new dataset that contains 48260 CT scan images from 282 normal persons and 15589 images from 95 patients with COVID-19 infections. At the first stage, this system runs our proposed image processing algorithm to discard those CT images that inside the lung is not properly visible in them. This action helps to reduce the processing time and false detections. At the next stage, we introduce a novel method for increasing the classification accuracy of convolutional networks. We implemented our method using the ResNet50V2 network and a modified feature pyramid network alongside our designed architecture for classifying the selected CT images into COVID-19 or normal with higher accuracy than other models. After running these two phases, the system determines the condition of the patient using a selected threshold. We are the first to evaluate our system in two different ways. In the single image classification stage, our model achieved 98.49% accuracy on more than 7996 test images. At the patient identification phase, the system correctly identified almost 234 of 245 patients with high speed. We also investigate the classified images with the Grad-CAM algorithm to indicate the area of infections in images and evaluate our model classification correctness.

Carbonate reservoirs, influenced by depositional and diagenetic processes and characterized by features like faults and vugs that impact storage capacity, require more than traditional Borehole Imaging logs (BHI) for accurate porosity data. This data is essential for geological assessments, production forecasting, and reservoir simulations. This work aims to address this limitation by developing methods to measure and monitor the sustainability of carbonate reservoirs and exploring the application of sustainability principles to their management. The study integrates BHI and conventional logs from two wells to classify porosity-based facies within the Barra Velha Formation (BVF) in the Santos Basin. The methodology involves four steps: (i) analyzing conventional logs; (ii) segmenting BHI logs; (iii) integrating conventional and segmented BHI logs using Self-Organizing Maps (SOM); and (iv) interpreting the resulting classes. Matrix porosity values and non-matrix pore sizes categorize the porosity into four facies (A to D). Facies A has high non-matrix porosity with 14,560 small megapores, 5,419 large megapores, and 271 gigapores (71.9%, 26.76%, and 1.34% of the 20,250 pores, respectively). Facies B shows moderate non-matrix porosity with 8,669 small megapores, 2,642 large megapores, and 33 gigapores (76.42%, 23.29%, and 0.29% of the 11,344 pores, respectively) and medium matrix porosity. Facies C exhibits low non-matrix porosity with 7,749 small megapores, 2,132 large megapores, and 20 gigapores (78.27%, 21.53%, and 0.20% of the 9,901 pores, respectively) and medium matrix porosity. Facies D has low non-matrix porosity with 9,355 small megapores, 2,346 large megapores, and 19 gigapores (79.82%, 20.02%, and 0.16% of the 11,720 pores, respectively) and low matrix porosity. The novelty of this work lies in integrating data from two sources to classify porosity across the Pre-Salt reservoir interval, serving as a proxy for preliminary litho-facies identification without core data. This study advances our understanding of carbonate reservoir sustainability and heterogeneity, offering valuable insights for robust, sustainable reservoir characterization and management in the context of global environmental and geological changes.

Breast cancer is the most common cause of death for women worldwide. Thus, the ability of artificial intelligence systems to predict and classify breast cancer is very important. In this paper, a hybrid ensemble method classification mechanism is proposed based on a majority voting mechanism. First, the performance of different state-of-the-art machine learning classification algorithms for the Wisconsin Breast Cancer Dataset (WBCD) were evaluated. The three best classifiers were then selected based on their F3 score. F3 score is used to emphasize the importance of false negatives (recall) in breast cancer classification. Then, these three classifiers, simple logistic Regression learning, stochastic gradient descent learning and multilayer perceptron network, are used for ensemble classification using a voting mechanism. We also evaluated the performance of hard and soft voting mechanism. For hard voting, majority-based voting mechanism was used and for soft voting we used average of probabilities, product of probabilities, maximum of probabilities and minimum of probabilities-based voting methods. The hard voting (majority-based voting) mechanism shows better performance with 99.42% as compared to the state-of-the-art algorithm for WBCD.

The last decade has seen a significant growth in low-cost air quality monitoring systems (LCAQMS) adoption, mostly driven by the spatial density limitations of traditional regulatory grade networks which restrict their effectiveness in several settings. The EU's regulatory framework advise for deploying one air quality monitoring unit per roughly 170,000 people or 15 square kilometers which is insufficient for capturing high spatio-temporal variance in complex urban scenarios. Often these settings are characterized by distributed emissions in small roads subjected to intense cars traffic, house heating driven emissions and frequently subjected to canyon effects. The Air Heritage project in Portici, funded by the EU's Urban Innovative Actions (UIA) program, addressed these challenges by integrating low-cost sensors with traditional monitoring systems and engaging the community in pervasive measurement strategies. The project highlights the importance of individualized sensor calibration and frequent recalibrations to maintain accuracy. It also showcases the benefits of citizen science approaches, IoT infrastructures, AI components, and geostatistical sensor fusion algorithms for mobile and opportunistic air quality measurements. This paper reports on the lessons learned and experiences from the Air Heritage project, aiming to guide future implementations in similar contexts. By sharing these insights along with the gathered datalake, we aim to inform stakeholders, including researchers, public authorities, and citizens, about effective strategies for deploying and utilizing LCAQMS to enhance air quality monitoring and public awareness.

Exercise-based cardiac rehabilitation requires patients to perform a set of certain prescribed exercises a specific number of times. Local muscular endurance (LME) exercises are an important part of the rehabilitation program. Automatic exercise recognition and repetition counting, from wearable sensor data is an important technology to enable patients to perform exercises independently in remote settings, e.g. their own home. In this paper we first report on a comparison of traditional approaches to exercise recognition and repetition counting, corresponding to supervised machine learning and peak detection from inertial sensing signals respectively, with more recent machine learning approaches, specifically Convolutional Neural Networks (CNNs). We investigated two different types of CNN: one using the AlexNet architecture, the other using time-series array. We found that the performance of CNN based approaches were better than the traditional approaches. For exercise recognition task, we found that the AlexNet based single CNN model outperformed other methods with an overall 97.18% F1-score measure. For exercise repetition counting , again the AlexNet architecture based single CNN model outperformed other methods by correctly counting repetitions in 90% of the performed exercise sets within an error of ±1. To the best of our knowledge, our approach of using a single CNN method for both recognition and repetition counting is novel. In addition to reporting our findings, we also make the dataset we created, the INSIGHT-LME dataset, publicly available to encourage further research.

This research presents a novel Customised Load Adaptive Framework (CLAF) for fault classifica-tion in Induction Motors (IMs), utilizing the Machinery Fault Prevention Technology (MFPT) Bearing Dataset. CLAF represents a pioneering approach that extends traditional fault classification methodologies by accounting for load variations and dataset customization. Through a meticulous two-phase process, it unveils load-dependent fault subclasses that have not been readily identified in traditional approaches. Additionally, new classes are created to accommodate the dataset's unique characteristics. Phase 1 involves the exploration of load-dependent patterns in time and frequency domain features using one-way Analysis of Variance (ANOVA) ranking and validation via bagged tree classifiers. In Phase 2, CLAF is applied to identify mild, moderate, and severe load-dependent fault subclasses through optimal Continuous Wavelet Transform (CWT) selection through Wavelet Singular Entropy (WSE) and CWT energy analysis. The results are compelling, with a 96.3% classification accuracy achieved when employing a wide neural network to classify proposed load-dependent fault subclasses. This underscores the practical value of CLAF in en-hancing fault diagnosis in IMs and its future potential in advancing IM condition monitoring.

Microarray gene expression-based detection and classification of medical conditions have been prominent in research studies over the past few decades. However, extracting relevant data from the high-volume microarray gene expression with inherent nonlinearity and inseparable noise components raises significant challenges during data classification and disease detection. So, this paper proposes a two-level strategy involving feature extraction and selection methods before the classification step. The feature extraction step utilizes Short Term Fourier Transform (STFT), and the feature selection step employs Particle Swarm Optimization (PSO) and Harmonic Search (HS) metaheuristic methods. The classifiers employed are Non-Linear Regression, Gaussian Mixture Model, Softmax Discriminant, Naive Bayes, SVM (Linear), SVM (Polynomial), and SVM (RBF). The two-level extracted relevant features are compared with raw data classification results, including Convolutional Neural Network (CNN) Methodology. Among the methods, STFT with PSO feature selection and SVM (RBF) classifier produced the highest accuracy of 94.47%.