Search

Fractional integrals and inequalities have gained a lot of attention in recent years. By introducing innovative analytical approaches and applications, and by applying these approaches, numerous forms of inequalities have been examined. In this paper, we establish new identity for the twice differentiable function where the absolute value is convex. By utilizing this identity, numerous Corrected Euler-Maclaurin-type inequalities are developed for the Caputo-Fabrizio fractional integral operator. Based on this identity, the Corrected Euler-Maclaurin-type inequalities for [Formula: see text]-convex function are obtained. By employing well-known inequalities such as Hölder’s and Power -Mean, we are introduced several new error bounds and estimates for Corrected Euler-Maclaurin-type inequalities. Additionally, special cases of the present results are applied to obtain the previous well-known results.

Image classification is an important application for deep learning. With the advent of quantum technology, quantum neural networks (QNNs) have become the focus of research. Traditional deep learning-based image classification involves using a convolutional neural network (CNN) to extract features from the image and a multi-layer perceptron (MLP) network to create the decision boundaries. However, quantum circuits with parameters can extract rich features from images and also create complex decision boundaries. This paper proposes a hybrid QNN (H-QNN) model designed for binary image classification that capitalizes on the strengths of quantum computing and classical neural networks. Our H-QNN model uses a compact, two-qubit quantum circuit integrated with a classical convolutional architecture, making it highly efficient for computation on noisy intermediate-scale quantum (NISQ) devices that are currently leading the way in practical quantum computing applications. Our H-QNN model significantly enhances classification accuracy, achieving a 90.1% accuracy rate on binary image datasets. In addition, we have extensively evaluated baseline CNN and our proposed H-QNN models for image retrieval tasks. The obtained quantitative results exhibit the generalization of our H-QNN for downstream image retrieval tasks. Furthermore, our model addresses the issue of overfitting for small datasets, making it a valuable tool for practical applications.

Proper nitrogen management in crops is crucial to ensure optimal growth and yield maximization. While hyperspectral imagery is often used for nitrogen status estimation in crops, it is not feasible for real-time applications due to the complexity and high cost associated with it. Much of the research utilizing RGB data for detecting nitrogen stress in plants relies on datasets obtained under laboratory settings, which limits its usability in practical applications. This study focuses on identifying nitrogen deficiency in maize crops using RGB imaging data from a publicly available dataset obtained under field conditions. We have proposed a custom-built vision transformer model for the classification of maize into three stress classes. Additionally, we have analyzed the performance of convolutional neural network models, including ResNet50, EfficientNetB0, InceptionV3, and DenseNet121, for nitrogen stress estimation. Our approach involves transfer learning with fine-tuning, adding layers tailored to our specific application. Our detailed analysis shows that while vision transformer models generalize well, they converge prematurely with a higher loss value, indicating the need for further optimization. In contrast, the fine-tuned CNN models classify the crop into stressed, non-stressed, and semi-stressed classes with higher accuracy, achieving a maximum accuracy of 97% with EfficientNetB0 as the base model. This makes our fine-tuned EfficientNetB0 model a suitable candidate for practical applications in nitrogen stress detection.

Situational awareness (SA) is of tremendous significance for successful operations in many domains, such as surveillance, humanitarian, search, and rescue missions, and national security. SA is particularly important for the defense sector, and is regarded as the decisive factor in military and air combat engagements. Commanders and operators rely on the accuracy and fidelity of SA for comprehending the environment, decision-making, and carrying out actions based on these decisions for accomplishing a mission. SA, however, is susceptible to adversarial attacks that can compromise the security and trust of SA systems. In this paper, we discuss the significance of security and trust of SA from an air force perspective. We then propose a model for quantifying the trustworthiness of an SA system. We further present numerical examples that demonstrate the quantification of trustworthiness of an SA system using our proposed model. Finally, we conclude this paper with future research directions for quantifying the security of SA systems.

To mitigate future educational challenges, the early childhood period is critical for cognitive development, so understanding the factors influencing child learning abilities is essential. This study investigates the impact of parenting techniques, sociodemographic characteristics, and health conditions on the learning abilities of children under five years old. Our primary goal is to explore the key factors that influence children’s learning abilities. For our study, we utilized the 2019 Multiple Indicator Cluster Surveys (MICS) dataset in Bangladesh. Using statistical analysis, we identified the key factors that affect children’s learning capability. To ensure proper analysis, we used extensive data preprocessing, feature manipulation, and model evaluation. Furthermore, we explored robust machine learning (ML) models to analyze and predict the learning challenges faced by children. These include logistic regression (LRC), decision tree (DT), k-nearest neighbor (KNN), random forest (RF), gradient boosting (GB), extreme gradient boosting (XGB), and bagging classification models. Out of these, GB and XGB, with 10-fold cross-validation, achieved an impressive accuracy of 95%, F1-score of 95%, and receiver operating characteristic area under the curve (ROC AUC) of 95%. Additionally, to interpret the model outputs and explore influencing factors, we used explainable AI (XAI) techniques like SHAP and LIME. Both statistical analysis and XAI interpretation revealed key factors that influence children’s learning difficulties. These include harsh disciplinary practices, low socioeconomic status, limited maternal education, and health-related issues. These findings offer valuable insights to guide policy measures to improve educational outcomes and promote holistic child development in Bangladesh and similar contexts.

Ensuring confidentiality and privacy is critical when it comes to sharing images over unsecured networks such as the internet. Since widely used and secure encryption methods, such as AES, Twofish, and RSA, are not suitable for real-time image encryption due to their slow encryption speeds and high computational requirements, researchers have proposed specialized algorithms for image encryption. This paper provides an introduction and overview of the image encryption algorithms and metrics used, aiming to evaluate them and help researchers and practitioners starting in this field obtain adequate information to understand the current state of image encryption algorithms. This paper classifies image encryption into seven different approaches based on the techniques used and analyzes the strengths and weaknesses of each approach. Furthermore, this paper provides a detailed review of a comprehensive set of security, quality, and efficiency evaluation metrics for image encryption algorithms, and provides upper and lower bounds for these evaluation metrics. Finally, this paper discusses the pros and cons of different image encryption approaches as well as the suitability of different image encryption approaches for different applications.

Quantum computing, a transformative field that emerged from quantum mechanics and computer science, has gained immense attention for its potential to revolutionize computation. This paper aims to address the fundamentals of quantum computing and provide a comprehensive guide for both novices and experts in the field of quantum computing. Beginning with the foundational principles of quantum computing, we introduce readers to the fundamental concepts of qubits, superposition, entanglement, interference, and noise. We explore quantum hardware, quantum gates, and basic quantum circuits. This study offers insight into the current phase of quantum computing, including the noisy intermediate-scale quantum (NISQ) era and its potential for solving real-world problems. Furthermore, we discuss the development of quantum algorithms and their applications, with a focus on famous algorithms like Shor’s algorithm and Grover’s algorithm. We also touch upon quantum computing’s impact on various industries, such as cryptography, optimization, machine learning, and material science. By the end of this paper, readers will have a solid understanding of quantum computing’s principles, applications, and the steps involved in developing quantum circuits. Our goal is to provide a valuable resource for those eager to embark on their quantum computing journey and for researchers looking to stay updated on this rapidly evolving field.

Blood smear analysis is often used to diagnose diseases like malaria, Anemia, Leukemia, etc. Morphological changes, such as size, shapes, and color, are receiving much attention in pathological analysis. Existing methods for detecting, diagnosing and analyzing blood smears cannot quantify overlapped, irregular boundaries and complex structures. This work proposes and evaluates a framework that utilizes Generative adversarial networks (GANs) for the segmentation and classification of blood elements, that is, white blood cells (WBCs), red blood cells (RBCs), and platelets (PLTs) simultaneously. The Generator of the network determines the mapping from microscopic images of blood cells to a confidence map. This mapping stipulates the probabilities of the pixel of the microscopic blood cell images with respect to ground truth. The Discriminator of the network is essential to castigate the mismatch between the microscopic blood cells images and confidence map. Additionally, adversarial learning enables the Generator to generate a qualitative confidence map that is converted into segmented images. We have calculated minimum, maximum, and average losses to judge the performance of the proposed model. We measure structural similarity, peak signal-to-noise ratio, pixel classification error, and finally, classified cells. The proposed framework can analyze all the blood cell elements simultaneously. The proposed framework shows a significant improvement in the segmentation and classification of blood cell elements compared to state-of-the-art techniques. During the training process, generator total loss reduces by 12.18%, 5.39%, and 3.62% for RBCs, WBCs, and PLTs, respectively. Our results demonstrate that the proposed framework outperforms existing state-of-the-art techniques, achieving the highest pixel correctly classified (PCC) ratio for the segmentation of blood cells as 99.8%, 93.4%, and 99.9% for WBCs, RBCs, and PLTs, respectively. Our framework attains 95.45% and 88.89% classification accuracy for WBCs on ALL-IDB-I and ALL-IDB-II datasets. The dataset used for this study can be found at https://drive.google.com/drive/folders/1F7kZ1SRWUD9R6aHLMkj3wsjcHnvlGuwP?usp=sharing

In the current era of digital communication, image security is a top priority as sending images, which often contain personal or confidential information, over the Internet poses security risks. Many encryption algorithms have been proposed in the literature to address this issue. However, these algorithms have limitations, such as low-key space, high computational overhead, and susceptibility to differential attacks. To address these limitations, this paper proposes Z-Crypt, a novel image encryption approach that combines a substitution-permutation network (SPN) and a chaotic logistic map (CLM) with the Chirp Z-Transform (CZT) to enhance security and resist attacks. The CLM-generated matrices introduce confusion and diffusion, while SPN creates a dissociation of the cipher from the plaintext. Finally, the CZT strengthens the encryption by transforming the image into the frequency domain, creating multiple layers of confusion and diffusion to produce a robust encryption algorithm. We have evaluated the proposed Z-Crypt’s security using various metrics, including correlation coefficient, entropy, peak signal-to-noise ratio (PSNR), and key sensitivity analysis. Experimental results verify that the proposed algorithm outperforms existing methods, achieving high security while maintaining computational efficiency.

Deep learning is employed in many applications, such as computer vision, natural language processing, robotics, and recommender systems. Large and complex neural networks lead to high accuracy; however, they adversely affect many aspects of deep learning performance, such as training time, latency, throughput, energy consumption, and memory usage in the training and inference stages. To solve these challenges, various optimization techniques and frameworks have been developed for the efficient performance of deep learning models in the training and inference stages. Although optimization techniques such as quantization have been studied thoroughly in the past, less work has been done to study the performance of frameworks that provide quantization techniques. In this paper, we have used different performance metrics to study the performance of various quantization frameworks, including TensorFlow automatic mixed precision and TensorRT. These performance metrics include training time and memory utilization in the training stage along with latency and throughput for graphics processing units (GPUs) in the inference stage. We have applied the automatic mixed precision (AMP) technique during the training stage using the TensorFlow framework, while for inference we have utilized the TensorRT framework for the post-training quantization technique using the TensorFlow TensorRT (TF-TRT) application programming interface (API).We performed model profiling for different deep learning models, datasets, image sizes, and batch sizes for both the training and inference stages, the results of which can help developers and researchers to devise and deploy efficient deep learning models for GPUs.

Introduction Achieving the Sustainable Development Goals to reduce maternal and neonatal mortality rates will require the expansion and strengthening of quality maternal health services. Midwife-led birth centres (MLBCs) are an alternative to hospital-based care for low-risk pregnancies where the lead professional at the time of birth is a trained midwife. These have been used in many countries to improve birth outcomes.Methods The cost analysis used primary data collection from four MLBCs in Bangladesh, Pakistan and Uganda (n=12 MLBC sites). Modelled cost-effectiveness analysis was conducted to compare the incremental cost-effectiveness ratio (ICER), measured as incremental cost per disability-adjusted life-year (DALY) averted, of MLBCs to standard care in each country. Results were presented in 2022 US dollars.Results Cost per birth in MLBCs varied greatly within and between countries, from US$21 per birth at site 3, Bangladesh to US$2374 at site 2, Uganda. Midwife salary and facility operation costs were the primary drivers of costs in most MLBCs. Six of the 12 MLBCs produced better health outcomes at a lower cost (dominated) compared with standard care; and three produced better health outcomes at a higher cost compared with standard care, with ICERs ranging from US$571/DALY averted to US$55 942/DALY averted.Conclusion MLBCs appear to be able to produce better health outcomes at lower cost or be highly cost-effective compared with standard care. Costs do vary across sites and settings, and so further exploration of costs and cost-effectiveness as a part of implementation and establishment activities should be a priority.

Fractional integral inequalities have become one of the most useful and expansive tools for the development of many fields of pure and applied mathematics over the past few years. Many authors have just recently introduced various generalized inequalities that involved the fractional integral operators. The main goal of the present study is to incorporate the concept of strongly $ \left(s, m\right) $-convex functions and Hermite-Hadamard inequality with Caputo-Fabrizio integral operator. Also, we consider a new identity for twice differentiable mapping in the context of Caputo-Fabrizio fractional integral operator. Then, considering this identity as an auxiliary result, new mid-point version using well known inequalities like Hölder, power-mean, Young are presented. Moreover, some graphs of obtained inequalities are given for better understanding by the reader. Finally, we discussed some applications to matrix inequalities and spacial means.

Scientific knowledge of image-based crack detection methods is limited in understanding their performance across diverse crack sizes, types, and environmental conditions. Builders and engineers often face difficulties with image resolution, detecting fine cracks, and differentiating between structural and non-structural issues. Enhanced algorithms and analysis techniques are needed for more accurate assessments. Hence, this research aims to generate an intelligent scheme that can recognize the presence of cracks and visualize the percentage of cracks from an image along with an explanation. The proposed method fuses features from concrete surface images through a ResNet-50 convolutional neural network (CNN) and curvelet transform handcrafted (HC) method, optimized by linear discriminant analysis (LDA), and the eXtreme gradient boosting (XGB) classifier then uses these features to recognize cracks. This study evaluates several CNN models, including VGG-16, VGG-19, Inception-V3, and ResNet-50, and various HC techniques, such as wavelet transform, counterlet transform, and curvelet transform for feature extraction. Principal component analysis (PCA) and LDA are assessed for feature optimization. For classification, XGB, random forest (RF), adaptive boosting (AdaBoost), and category boosting (CatBoost) are tested. To isolate and quantify the crack region, this research combines image thresholding, morphological operations, and contour detection with the convex hulls method and forms a novel algorithm. Two explainable AI (XAI) tools, local interpretable model-agnostic explanations (LIMEs) and gradient-weighted class activation mapping++ (Grad-CAM++) are integrated with the proposed method to enhance result clarity. This research introduces a novel feature fusion approach that enhances crack detection accuracy and interpretability. The method demonstrates superior performance by achieving 99.93% and 99.69% accuracy on two existing datasets, outperforming state-of-the-art methods. Additionally, the development of an algorithm for isolating and quantifying crack regions represents a significant advancement in image processing for structural analysis. The proposed approach provides a robust and reliable tool for real-time crack detection and assessment in concrete structures, facilitating timely maintenance and improving structural safety. By offering detailed explanations of the model’s decisions, the research addresses the critical need for transparency in AI applications, thus increasing trust and adoption in engineering practice.

As recent machine translation models are mostly based on the attention-based neural machine translation (NMT), many well-known models such as Transformer or bidirectional encoder representations from Transformers (BERT) have been proposed. Along with algorithmic advancements, hardware acceleration methods for those attention-based neural machine translation models have also been introduced. However, the size of the parameters for attention-based NMT is also becoming larger to guarantee the satisfactory machine translation quality. Among various weights, linearization weights ( $W^{Q}$ , $W^{K}$ , $W^{V}$ , and $W^{O}$ ) account for a non-negligible portion (by up to 30%) among the entire parameters in the modern NMT models. In this paper, we propose a method for linearization weight compression and near-memory hardware decoder for fast and in-situ weight decompression. Our weight compression method exploits the fixed-point quantization along with Huffman coding which is selectively applied depending on the weight value distribution. Our hardware decoder decompresses the Huffman-coded weights near-memory to minimize the weight decoding latency. Our compression method shows 4.9–10.0 compression ratio with small NMT score drops across the five widely used attention-based NMT models (Transformer, Transformer-XL-base, Transformer-XL-large, BERT-base, and BERT-large). In addition, due to the reduced linearization weight size, our proposed method with near-memory decoding enables multi-head attention (MHA) execution latency reduction by 11.8%, on average, as compared to the baseline when considering the weight loading and initialization. In terms of the memory data transfer energy consumption, our proposed method leads to a memory energy saving of 16.1%, on average, as compared to the baseline.

Human activity recognition is an emerging and important area in computer vision which seeks to determine the activity an individual or group of individuals are performing. The applications of this field ranges from generating highlight videos in sports, to intelligent surveillance and gesture recognition. Most activity recognition systems rely on a combination of convolutional neural networks (CNNs) to perform feature extraction from the data and recurrent neural networks (RNNs) to determine the time dependent nature of the data. This paper proposes and designs two transformer neural networks for human activity recognition: a recurrent transformer (ReT), a specialized neural network used to make predictions on sequences of data, as well as a vision transformer (ViT), a transformer optimized for extracting salient features from images, to improve speed and scalability of activity recognition. We have provided an extensive comparison of the proposed transformer neural networks with the contemporary CNN and RNN-based human activity recognition models in terms of speed and accuracy for four publicly available human action datasets. Experimental results reveal that the proposed ViT-ReT framework attains a speedup of $2\times $ over the baseline ResNet50-LSTM approach while attaining nearly the same level of accuracy. Furthermore, results show that the proposed ViT-ReT framework attains significant improvements over the state-of-the-art human action recognition methods in terms of both model accuracy and runtime for each of the datasets used in our experiments, thus verifying the suitability of the proposed ViT-ReT framework for human activity recognition in resource-constrained and real-time environments.

Transferring an all-electric spacecraft from a launch injection orbit to the geosynchronous equatorial orbit (GEO) using a low thrust propulsion system presents a significant challenge due to the long transfer time typically spanning several months. To address the challenge of determining such long time-scale orbit-raising maneuvers to GEO, this paper presents a novel technique to compute transfers starting from geostationary transfer orbit (GTO) and super-GTO. The transfer is complex, involving multiple eclipses and revolutions. To tackle this challenge, we introduce a cascaded deep reinforcement learning (DRL) model to guide a low-thrust spacecraft towards the desired orbit by determining an appropriate thrust direction at each state. To ensure mission requirements, a gradient-aided reward function incorporating the orbital elements, guides the DRL agent to obtain the optimal flight time. The obtained results demonstrate that our proposed approach yields optimal or near-optimal time-efficient spacecraft orbit-raising. DRL implementation is important for spacecraft autonomy; in this context, we demonstrate that our DRL-based trajectory planning provides significantly better transfer time as compared to state-of-the-art approaches that allow for automated trajectory computation.

Fractional calculus has been a concept used to obtain new variants of some well-known integral inequalities. In this study, our main goal is to establish the new fractional Hermite–Hadamard, and Simpson’s type estimates by employing a differentiable function. Furthermore, a novel class of fractional integral related to prominent fractional operator (Caputo–Fabrizio) for differentiable convex functions of first order is proven. Then, taking this equality into account as an auxiliary result, some new estimation of the Hermite–Hadamard and Simpson’s type inequalities as generalization is presented. Moreover, few inequalities for concave function are obtained as well. It is observed that newly established outcomes are the extension of comparable inequalities existing in the literature. Additionally, we discuss the applications to special means, matrix inequalities, and the q-digamma function.

To obtain more consistent measurements through the course of a wheat growing season, we conceived and designed an autonomous robotic platform that performs collision avoidance while navigating in crop rows using spatial artificial intelligence (AI). The main constraint the agronomists have is to not run over the wheat while driving. Accordingly, we have trained a spatial deep learning model that helps navigate the robot autonomously in the field while avoiding collisions with the wheat. To train this model, we used publicly available databases of prelabeled images of wheat, along with the images of wheat that we have collected in the field. We used the MobileNet single shot detector (SSD) as our deep learning model to detect wheat in the field. To increase the frame rate for real-time robot response to field environments, we trained MobileNet SSD on the wheat images and used a new stereo camera, the Luxonis Depth AI Camera. Together, the newly trained model and camera could achieve a frame rate of 18–23 frames per second (fps)—fast enough for the robot to process its surroundings once every 2–3 inches of driving. Once we knew the robot accurately detects its surroundings, we addressed the autonomous navigation of the robot. The new stereo camera allows the robot to determine its distance from the trained objects. In this work, we also developed a navigation and collision avoidance algorithm that utilizes this distance information to help the robot see its surroundings and maneuver in the field, thereby precisely avoiding collisions with the wheat crop. Extensive experiments were conducted to evaluate the performance of our proposed method. We also compared the quantitative results obtained by our proposed MobileNet SSD model with those of other state-of-the-art object detection models, such as the YOLO V5 and Faster region-based convolutional neural network (R-CNN) models. The detailed comparative analysis reveals the effectiveness of our method in terms of both model precision and inference speed.

Convexity plays a crucial role in the development of fractional integral inequalities. Many fractional integral inequalities are derived based on convexity properties and techniques. These inequalities have several applications in different fields such as optimization, mathematical modeling and signal processing. The main goal of this article is to establish a novel and generalized identity for the Caputo–Fabrizio fractional operator. With the help of this specific developed identity, we derive new fractional integral inequalities via exponential convex functions. Furthermore, we give an application to some special means.

The Internet of Things (IoT) is a mesh network of interconnected objects with unique identifiers that can transmit data and communicate with one another without the need for human intervention. The IoT has brought the future closer to us. It has opened up new and vast domains for connecting not only people, but also all kinds of simple objects and phenomena all around us. With billions of heterogeneous devices connected to the Internet, the network architecture must evolve to accommodate the expected increase in data generation while also improving the security and efficiency of connectivity. Traditional IoT architectures are primitive and incapable of extending functionality and productivity to the IoT infrastructure’s desired levels. Software-Defined Networking (SDN) and virtualization are two promising technologies for cost-effectively handling the scale and versatility required for IoT. In this paper, we discussed traditional IoT networks and the need for SDN and Network Function Virtualization (NFV), followed by an analysis of SDN and NFV solutions for implementing IoT in various ways.

Privacy is a serious concern related to sharing videos or images among people over the Internet. As a method to preserve images’ privacy, chaos-based image encryption algorithms have been used widely to fulfil such a requirement. However, these algorithms suffer from a low key-space, significant computational overhead, and a lag in resistance against differential attacks. This paper presents a novel chaos-based image encryption method based on permutation and substitution using a single Substitution Box (S-Box) to address issues in contemporary image encryption algorithms. The proposed encryption technique’s efficiency is validated through extensive experiments as compared to the state-of-the-art encryption algorithms using different measures and benchmarks. Precisely, the collected results demonstrate that the proposed technique is more resilient against well-known statistical attacks and performs well under plaintext attacks. Indeed, the proposed scheme exhibits very high sensitivity concerning the plaintext attack. A minor change in the encryption key or the plain text would result in a completely different encrypted image.

Matrix multiplication is a main computation kernel of emerging workloads, such as deep neural networks and graph analytics. These workloads often exhibit high sparsity in data, which means a large portion of the elements in the data are zero-valued elements. Though systolic arrays have shown a significant performance and energy efficiency improvement over central processing units (CPUs) or graphic processing units (GPUs) when executing matrix multiplications, data sparsity is largely overlooked in the conventional systolic arrays. In this paper, we propose a row-wise product-based sparse matrix multiplication (SpMM) hardware accelerator for compressed sparse row (CSR)-formatted input matrices. Our hardware accelerator leverages row-wise product, which has advantages over inner-product or outer-product when executing the sparse matrix multiplications. As compared to the conventional systolic arrays, which cannot skip the ineffectual operations, our hardware accelerator only performs effectual operations with non-zero elements, improving the performance when executing SpMM. In addition, we also propose an optimal load balancing scheme when using multiple processing elements (PEs). Our load balancing scheme utilizes an operation count-based matrix tiling for parallel execution of the PEs and resource contention avoidance. According to our evaluation, our 32PE-SpMM accelerator shows $13.6\times $ – $47.9\times $ speedup over tensor processing unit (TPU)-like systolic arrays, on average. Furthermore, our operation count-based load balancing scheme shows better performance over the fixed tiling and non-zero element count-based tiling by up to 8.48% and 6.28%, respectively, with only up to 0.06% matrix tiling pre-processing latency overhead.

Situational awareness (SA) is defined as the perception of entities in the environment, comprehension of their meaning, and projection of their status in near future. From an Air Force perspective, SA refers to the capability to comprehend and project the current and future disposition of red and blue aircraft and surface threats within an airspace. In this article, we propose a model for SA and dynamic decision-making that incorporates artificial intelligence and dynamic data-driven application systems to adapt measurements and resources in accordance with changing situations. We discuss measurement of SA and the challenges associated with quantification of SA. We then elaborate a plethora of techniques and technologies that help improve SA ranging from different modes of intelligence gathering to artificial intelligence to automated vision systems. We then present different application domains of SA including battlefield, gray zone warfare, military- and air-base, homeland security and defense, and critical infrastructure. Finally, we conclude the article with insights into the future of SA.

Abstract Many recent research efforts have exploited data sparsity for the acceleration of convolutional neural network (CNN) inferences. However, the effects of data transfer between main memory and the CNN accelerator have been largely overlooked. In this work, the authors propose a CNN acceleration technique that leverages hardware/software co‐design and exploits the sparsity in input feature maps (IFMs). On the software side, the authors' technique employs a novel lossless compression scheme for IFMs, which are sent to the hardware accelerator via direct memory access. On the hardware side, the authors' technique uses a CNN inference accelerator that performs convolutional layer operations with their compressed data format. With several design optimization techniques, the authors have implemented their technique in a field‐programmable gate array (FPGA) system‐on‐chip platform and evaluated their technique for six different convolutional layers in SqueezeNet. Results reveal that the authors' technique improves the performance by 1.1×–22.6× while reducing energy consumption by 47.7%–97.4% as compared to the CPU‐based execution. Furthermore, results indicate that the IFM size and transfer latency are reduced by 34.0%–85.2% and 4.4%–75.7%, respectively, compared to the case without data compression. In addition, the authors' hardware accelerator shows better performance per hardware resource with less than or comparable power consumption to the state‐of‐the‐art FPGA‐based designs.

With the exponential rise in government and private health-supported schemes, the number of fraudulent billing cases is also increasing. Detection of fraudulent transactions in healthcare systems is an exigent task due to intricate relationships among dynamic elements, including doctors, patients, and services. Hence, to introduce transparency in health support programs, there is a need to develop intelligent fraud detection models for tracing the loopholes in existing procedures, so that the fraudulent medical billing cases can be accurately identified. Moreover, there is also a need to optimize both the cost burden for the service provider and medical benefits for the client. This paper presents a novel process-based fraud detection methodology to detect insurance claim-related frauds in the healthcare system using sequence mining concepts. Recent literature focuses on the amount-based analysis or medication versus disease sequential analysis rather than detecting frauds using sequence generation of services within each specialty. The proposed methodology generates frequent sequences with different pattern lengths. The confidence values and confidence level are computed for each sequence. The sequence rule engine generates frequent sequences along with confidence values for each hospital’s specialty and compares them with the actual patient values. This identifies anomalies as both sequences would not be compliant with the rule engine’s sequences. The process-based fraud detection methodology is validated using last five years of a local hospital’s transactional data that includes many reported cases of fraudulent activities.

Recently, there has been a surge in use of electric propulsion to transfer satellites to the geostationary Earth orbit (GEO). Traditionally, the transfer times to reach GEO using all-electric propulsion are obtained by solving challenging trajectory optimization problems that naturally do not lend themselves to incorporation within deep reinforcement learning (DRL) framework to solve trajectory planning problems in near real-time. The operation of DRL, as typically used in trajectory planning, relies on a Q-value. In the electric orbit-raising problem under consideration in this paper, this Q-Value requires computation of transfer time in near real-time to have practical DRL training times. This work proposes to design and evaluate a machine learning (ML) framework, focusing on deep neural networks (DNNs), to predict the transfer time to assist in Q-value determination instead of solving traditional orbit-raising optimization problems. To this end, we investigate different architectures for DNNs to determine a suitable DNN configuration that can predict the transfer time for each of the mission scenarios with high accuracy. Experimental results indicate that our designed DNNs can predict the transfer time for different scenarios with an accuracy of over 99.97%. To verify the efficacy of our designed DNNs for predicting transfer time that is required for Q-value estimation, we also compare the results from our designed DNNs with the contemporary ML algorithms, such as support vector machines, random forests, and decision trees for regression. Experimental results indicate that our best-performing DNNs can provide an improvement in the mean error of transfer time prediction by up to 14.05× for non-planar transfers and up to 254× for planar transfers as compared to contemporary ML algorithms.

The recognition of human activities using vision-based techniques has become a crucial research field in video analytics. Over the last decade, there have been numerous advancements in deep learning algorithms aimed at accurately detecting complex human actions in video streams. While these algorithms have demonstrated impressive performance in activity recognition, they often exhibit a bias towards either model performance or computational efficiency. This biased trade-off between robustness and efficiency poses challenges when addressing complex human activity recognition problems. To address this issue, this paper presents a computationally efficient yet robust approach, exploiting saliency-aware spatial and temporal features for human action recognition in videos. To achieve effective representation of human actions, we propose an efficient approach called the dual-attentional Residual 3D Convolutional Neural Network (DA-R3DCNN). Our proposed method utilizes a unified channel-spatial attention mechanism, allowing it to efficiently extract significant human-centric features from video frames. By combining dual channel-spatial attention layers with residual 3D convolution layers, the network becomes more discerning in capturing spatial receptive fields containing objects within the feature maps. To assess the effectiveness and robustness of our proposed method, we have conducted extensive experiments on four well-established benchmark datasets for human action recognition. The quantitative results obtained validate the efficiency of our method, showcasing significant improvements in accuracy of up to 11% as compared to state-of-the-art human action recognition methods. Additionally, our evaluation of inference time reveals that the proposed method achieves up to a 74× improvement in frames per second (FPS) compared to existing approaches, thus showing the suitability and effectiveness of the proposed DA-R3DCNN for real-time human activity recognition.

Vision-based human activity recognition (HAR) has emerged as one of the essential research areas in video analytics. Over the last decade, numerous advanced deep learning algorithms have been introduced to recognize complex human actions from video streams. These deep learning algorithms have shown impressive performance for the video analytics task. However, these newly introduced methods either exclusively focus on model performance or the effectiveness of these models in terms of computational efficiency, resulting in a biased trade-off between robustness and computational efficiency in their proposed methods to deal with challenging HAR problem. To enhance both the accuracy and computational efficiency, this paper presents a computationally efficient yet generic spatial–temporal cascaded framework that exploits the deep discriminative spatial and temporal features for HAR. For efficient representation of human actions, we propose an efficient dual attentional convolutional neural network (DA-CNN) architecture that leverages a unified channel–spatial attention mechanism to extract human-centric salient features in video frames. The dual channel–spatial attention layers together with the convolutional layers learn to be more selective in the spatial receptive fields having objects within the feature maps. The extracted discriminative salient features are then forwarded to a stacked bi-directional gated recurrent unit (Bi-GRU) for long-term temporal modeling and recognition of human actions using both forward and backward pass gradient learning. Extensive experiments are conducted on three publicly available human action datasets, where the obtained results verify the effectiveness of our proposed framework (DA-CNN+Bi-GRU) over the state-of-the-art methods in terms of model accuracy and inference runtime across each dataset. Experimental results show that the DA-CNN+Bi-GRU framework attains an improvement in execution time up to 167× in terms of frames per second as compared to most of the contemporary action-recognition methods.

Urban surveillance, of which airborne urban surveillance is a vital constituent, provides situational awareness (SA) and timely response to emergencies. The significance and scope of urban surveillance has increased manyfold in recent years due to the proliferation of unmanned aerial vehicles (UAVs), Internet of things (IoTs), and multitude of sensors. In this article, we propose FogSurv—a fog-assisted surveillance architecture and framework leveraging artificial intelligence (AI) and information/data fusion for enabling real-time SA and monitoring. We also propose an AI- and data-driven information fusion model for FogSurv to help provide (near) real-time SA, threat assessment, and automated decision-making. We further present a latency model for AI and information fusion processing in FogSurv. We then discuss several use cases of FogSurv that can have a huge impact on multifarious fronts of national significance ranging from safeguarding national security to monitoring of critical infrastructures. We conduct an extensive set of experiments to demonstrate that FogSurv using AI and data fusion help provide near real-time inferences and SA. Experimental results demonstrate that FogSurv provides a latency improvement of 37% on average over cloud architectures for the selected benchmarks. Results further indicate that combining AI with data fusion as in FogSurv can provide a speedup of up to $9.8\times $ over AI without data fusion while also maintaining or improving the inference accuracy. Additionally, results show that AI combined with fusion of different image modalities obtained through UAVs in FogSurv results in improved average precision of target detection for surveillance as compared to AI without data fusion for different target scales and environment complexity.

Convolutional neural networks (CNNs) have gained a huge attention for real-world artificial intelligence (AI) applications such as image classification and object detection. On the other hand, for better accuracy, the size of the CNNs’ parameters (weights) has been increasing, which in turn makes it difficult to enable on-device CNN inferences in resource-constrained edge devices. Though weight pruning and 5-bit quantization methods have shown promising results, it is still challenging to deploy large CNN models in edge devices. In this paper, we propose an encoding and hardware-based decoding technique which can be applied to 5-bit quantized weight data for on-device CNN inferences in resource-constrained edge devices. Given 5-bit quantized weight data, we employ arithmetic coding with range scaling for lossless weight compression, which is performed offline. When executing on-device inferences with underlying CNN accelerators, our hardware decoder enables a fast in-situ weight decompression with small latency overhead. According to our evaluation results with five widely used CNN models, our arithmetic coding-based encoding method applied to 5-bit quantized weights shows a better compression ratio by $9.6\times $ while also reducing the memory data transfer energy consumption by 89.2%, on average, as compared to the case of uncompressed 32-bit floating-point weights. When applying our technique to pruned weights, we obtain better compression ratios by $57.5\times $ – $112.2\times $ while reducing energy consumption by 98.3%–99.1% as compared to the case of 32-bit floating-point weights. In addition, by pipelining the weight decoding and transfer with the CNN execution, the latency overhead of our weight decoding with 16 decoding unit (DU) hardware is only 0.16%–5.48% and 0.16%–0.91% for non-pruned and pruned weights, respectively. Moreover, our proposed technique with 4-DU decoder hardware reduces system-level energy consumption by 1.1%–9.3%.

Sparse convolutional neural network (CNN) models reduce the massive compute and memory bandwidth requirements inherently present in dense CNNs without a significant loss in accuracy. Sparse CNNs, however, present their own set of challenges including non-linear data accesses and complex design of CNN processing elements (PEs). Recently proposed accelerators like SCNN, Eyeriss v2, and SparTen, exploit the two-sided sparsity, that is, sparsity in both the input activations and weights to accelerate the CNN inference. These, accelerators, however, suffer from a multitude of problems that limit their applicability, such as inefficient micro-architecture (SCNN, Eyeriss v2), complex PE design (Eyeriss v2), no support for non-unit stride convolutions (SCNN) and FC layers (SparTen, SCNN). To address these issues in contemporary sparse CNN accelerators, we propose Sparse-PE, a multi-threaded, and flexible CNN PE, capable of handling both the dense and sparse CNNs. The Sparse-PE core uses binary mask representation and actively skips computations involving zeros and favors non-zero computations, thereby, drastically increasing the effective throughput and hardware utilization. Unlike previous designs, the Sparse-PE core is generic in nature and not targeted towards a specific accelerator, and thus, can also be used as a standalone sparse dot product compute engine. We evaluate the performance of the core using a custom built cycle accurate simulator. Our simulations show that the Sparse-PE core-based accelerator provides a performance gain of $12\times $ over a recently proposed dense accelerator (NeuroMAX). For sparse accelerators, it provides a performance gain of $4.2\times $ , $2.38\times $ , and $1.98\times $ over SCNN, Eyeriss v2, and SparTen, respectively.

Human action recognition has been actively explored over the past two decades to further advancements in video analytics domain. Numerous research studies have been conducted to investigate the complex sequential patterns of human actions in video streams. In this paper, we propose a knowledge distillation framework, which distills spatio-temporal knowledge from a large teacher model to a lightweight student model using an offline knowledge distillation technique. The proposed offline knowledge distillation framework takes two models: a large pre-trained 3DCNN (three-dimensional convolutional neural network) teacher model and a lightweight 3DCNN student model (i.e., the teacher model is pre-trained on the same dataset on which the student model is to be trained on). During offline knowledge distillation training, the distillation algorithm trains only the student model to help enable the student model to achieve the same level of prediction accuracy as the teacher model. To evaluate the performance of the proposed method, we conduct extensive experiments on four benchmark human action datasets. The obtained quantitative results verify the efficiency and robustness of the proposed method over the state-of-the-art human action recognition methods by obtaining up to 35% improvement in accuracy over existing methods. Furthermore, we evaluate the inference time of the proposed method and compare the obtained results with the inference time of the state-of-the-art methods. Experimental results reveal that the proposed method attains an improvement of up to 50× in terms of frames per seconds (FPS) over the state-of-the-art methods. The short inference time and high accuracy make our proposed framework suitable for human activity recognition in real-time applications.

The magnetic levitation (MAGLEV) train uses magnetic field to suspend, guide, and propel vehicle onto the track. The MAGLEV train provides a sustainable and cleaner solution for train transportation by significantly reducing the energy usage and greenhouse gas emissions as compared to traditional train transportation systems. In this paper, we propose an advanced control mechanism using an Arduino microcontroller that selectively energizes the electromagnets in a MAGLEV train system to provide dynamic stability and energy efficiency. We also design the prototype of an energy-efficient MAGLEV train that leverages our proposed control mechanism. In our MAGLEV train prototype, the levitation is achieved by creating a repulsive magnetic field between the train and the track using magnets mounted on the top-side of the track and bottom-side of the vehicle. The propulsion is performed by creating a repulsive magnetic field between the permanent magnets attached on the sides of the vehicle and electromagnets mounted at the center of the track using electrodynamic suspension (EDS). The electromagnets are energized via a control mechanism that is applied through an Arduino microcontroller. The Arduino microcontroller is programmed in such a way to propel and guide the vehicle onto the track by appropriate switching of the electromagnets. We use an infrared-based remote-control device for controlling the power, speed, and direction of the vehicle in both the forward and the backward direction. The proposed MAGLEV train control mechanism is novel, and according to the best of our knowledge is the first study of its kind that uses an Arduino-based microcontroller system for control mechanism. Experimental results illustrate that the designed prototype consumes only 144 W-hour (Wh) of energy as compared to a conventionally designed MAGLEV train prototype that consumes 1200 Wh. Results reveal that our proposed control mechanism and prototype model can reduce the total power consumption by 8.3 × as compared to the traditional MAGLEV train prototype, and can be applied to practical MAGLEV trains with necessary modifications. Thus, our proposed prototype and control mechanism serves as a first step towards cleaner engineering of train transportation systems.

Convolutional neural networks (CNNs) are widely deployed for many artificial intelligence (AI) applications, such as object detection and image classification. Due to the burgeoning revolution in edge AI, CNN hardware accelerators are also being employed in resource-constrained edge devices for achieving better performance and energy efficiency at the edge. Although CNN accelerators enable fast and energy-efficient CNN inference at the edge, the remaining hardware resources on the edge devices except for the CNN accelerator remain idle, which could otherwise be utilized for attaining even better performance and energy efficiency for CNN inferences. In this paper, we propose a CPU-accelerator co-scheduling technique for convolution (CONV) layer operations of CNN inferences in resource-constrained edge devices. Our proposed co-scheduling technique exploits an inherent parallelism in CNN output channels, that is, the operations for generating different output channels in a CONV layer can be executed in parallel. For load balancing between the CPU and the CNN accelerator, we also propose a simple, yet accurate latency model for CONV layer operations in the CPU and the accelerator. Based on the latency estimation of CONV layer operations provided by our proposed model, we distribute the tasks to the CPU and the CNN accelerator in a load-balance manner to minimize the idle period during the CONV layer operations in both the CPU and the CNN accelerator. We implement our proposed hardware/software (HW/SW) co-scheduling technique in various field-programmable gate array system-on-chip (FPGA-SoC) platforms as a proof-of-concept. Experimental results indicate that our proposed co-scheduling technique improves system performance by 1.18×-2.00× with energy reduction of 14.9% - 49.7% as compared to the accelerator-only execution.

The internet of things (IoT) is prone to various types of denial of service (DoS) attacks due to their resource-constrained nature. Extensive research efforts have been dedicated to securing these systems, but various vulnerabilities remain. Notably, it is challenging to maintain the confidentiality, integrity, and availability of mobile ad hoc networks due to limited connectivity and dynamic topology. As critical infrastructure including smart grids, industrial control, and intelligent transportation systems is reliant on WSNs and IoT, research efforts that forensically investigate and analyze the cybercrimes in IoT and WSNs are imperative. When a security failure occurs, the causes, vulnerabilities, and facts behind the failure need to be revealed and examined to improve the security of these systems. This research forensically investigates the performance of the ad hoc IoT networks using the ad hoc on-demand distance vector (AODV) routing protocol under the blackhole attack, which is a type of denial of service attack detrimental to IoT networks. This work also examines the traffic patterns in the network and nodes to assess the attack damage and conducts vulnerability analysis of the protocol to carry out digital forensic (DF) investigations. It further reconstructs the networks under different modes and parameters to verify the analysis and provide suggestions to design roubust routing protocols.

Chaotic-maps-based image encryption methods have been a topic of research interest for a decade. However, most of the proposed methods suffer from slow encryption time or compromise on the security of the encryption to achieve faster encryption. This paper proposes a lightweight, secure, and efficient image encryption algorithm based on logistic map, permutations, and AES S-box. In the proposed algorithm, SHA-2 based on the plaintext image, a pre-shared key, and an initialization vector (IV) are used to generate the initial parameters for the logistic map. The logistic map chaotically generates random numbers, which are then used for the permutations and substitutions. The security, quality, and efficiency of the proposed algorithm are tested and analyzed using a number of metrics, such as correlation coefficient, chi-square, entropy, mean square error, mean absolute error, peak signal-to-noise ratio, maximum deviation, irregular deviation, deviation from uniform histogram, number of pixel change rate, unified average changing intensity, resistance to noise and data loss attacks, homogeneity, contrast, energy, and key space and key sensitivity analysis. Experimental results reveal that the proposed algorithm is up to 15.33× faster compared to other contemporary encryption methods.

The Internet of things (IoT) enables a diverse set of applications such as distribution automation, smart cities, wireless sensor networks, and advanced metering infrastructure (AMI). In smart grids (SGs), quality of service (QoS) and AMI traffic management need to be considered in the design of efficient AMI architectures. In this article, we propose a QoS-aware machine-learning-based framework for AMI applications in smart grids. Our proposed framework comprises a three-tier hierarchical architecture for AMI applications, a machine-learning-based hierarchical clustering approach, and a priority-based scheduling technique to ensure QoS in AMI applications in smart grids. We introduce a three-tier hierarchical architecture for AMI applications in smart grids to take advantage of IoT communication technologies and the cloud infrastructure. In this architecture, smart meters are deployed over a georeferenced area where the control center has remote access over the Internet to these network devices. More specifically, these devices can be digitally controlled and monitored using simple web interfaces such as REST APIs. We modify the existing K-means algorithm to construct a hierarchical clustering topology that employs Wi-SUN technology for bi-directional communication between smart meters and data concentrators. Further, we develop a queuing model in which different priorities are assigned to each item of the critical and normal AMI traffic based on its latency and packet size. The critical AMI traffic is scheduled first using priority-based scheduling while the normal traffic is scheduled with a first-in–first-out scheduling scheme to ensure the QoS requirements of both traffic classes in the smart grid network. The numerical results demonstrate that the target coverage and connectivity requirements of all smart meters are fulfilled with the least number of data concentrators in the design. Additionally, the numerical results show that the architectural cost is reduced, and the bottleneck problem of the data concentrator is eliminated. Furthermore, the performance of the proposed framework is evaluated and validated on the CloudSim simulator. The simulation results of our proposed framework show efficient performance in terms of CPU utilization compared to a traditional framework that uses single-hop communication from smart meters to data concentrators with a first-in–first-out scheduling scheme.

The elliptic curve cryptosystem (ECC) has been proven to be vulnerable to non-invasive side-channel analysis attacks, such as timing, power, visible light, electromagnetic emanation, and acoustic analysis attacks. In ECC, the scalar multiplication component is considered to be highly susceptible to side-channel attacks (SCAs) because it consumes the most power and leaks the most information. In this work, we design a robust asynchronous circuit for scalar multiplication that is resistant to state-of-the-art timing, power, and fault analysis attacks. We leverage the genetic algorithm with multi-objective fitness function to generate a standard Boolean logic-based combinational circuit for scalar multiplication. We transform this circuit into a multi-threshold dual-spacer dual-rail delay-insensitive logic (MTD3L) circuit. We then design point-addition and point-doubling circuits using the same procedure. Finally, we integrate these components together into a complete secure and dependable ECC processor. We design and validate the ECC processor using Xilinx ISE 14.7 and implement it in a Xilinx Kintex-7 field-programmable gate array (FPGA).