Search Results (5,946)

The demand for fast and accurate electromagnetic solutions to support current and emerging technologies has fueled the rapid development of various machine learning techniques for applications such as antenna design and optimization, microwave imaging, device diagnostics, and more. Multi-fidelity (MF) surrogate modeling methods have shown great promise in significantly reducing computational costs associated with surrogate modeling while maintaining high model accuracy. This work offers a comprehensive review of the available MF surrogate modeling methods in electromagnetics, focusing on specific methodologies, related challenges, and the generation of variable-fidelity datasets. The article is structured around the two main types of electromagnetic problems: forward and inverse. It begins by summarizing key machine learning concepts and limitations. This transitions to discussing multi-fidelity surrogate model architectures and low-fidelity data techniques for the forward problem. Subsequently, the unique challenges of the inverse problem are presented, along with traditional solutions and their limitations. Following this, the review examines MF surrogate modeling approaches tailored to the inverse problem. In conclusion, the review outlines promising future directions in MF modeling for electromagnetics, aiming to provide fundamental insights into understanding these developing methods. Full article

Developing reliable railway fault detection systems is crucial for ensuring both safety and operational efficiency. Various artificial intelligence frameworks, especially deep learning models, have shown significant potential in enhancing fault detection within railway infrastructure. This study explores the application of deep learning models for railway fault detection, focusing on both transfer learning architectures and a novel classification framework. Transfer learning was utilized with architectures such as ResNet50V2, Xception, VGG16, MobileNet, and InceptionV3, which were fine-tuned to classify railway track images into defective and non-defective categories. Additionally, the state-of-the-art YOLOv11 model was adapted for the same classification task, leveraging advanced data augmentation techniques to achieve high accuracy. Among the transfer learning models, VGG16 demonstrated the best performance with a test accuracy of 89.18%. However, YOLOv11 surpassed all models, achieving a test accuracy of 92.64% while maintaining significantly lower computational demands. These findings underscore the versatility of deep learning models and highlight the potential of YOLOv11 as an efficient and accurate solution for railway fault classification tasks. Full article

To meet the high-reliability requirements of real-time on-orbit tasks, this paper proposes a fault-tolerant reinforcement design method for spaceborne intelligent processing algorithms based on convolutional neural networks (CNNs). This method is built on a CNN accelerator using Field-Programmable Gate Array (FPGA) technology, analyzing the impact of Single-Event Upsets (SEUs) on neural network computation. The accelerator design integrates data validation, Triple Modular Redundancy (TMR), and other techniques, optimizing a partial fault-tolerant architecture based on SEU sensitivity. This fault-tolerant architecture analyzes the hardware accelerator, parameter storage, and actual computation, employing data validation to reinforce model parameters and spatial and temporal TMR to reinforce accelerator computations. Using the ResNet18 model, fault tolerance performance tests were conducted by simulating SEUs. Compared to the prototype network, this fault-tolerant design method increases tolerance to SEU error accumulation by five times while increasing resource consumption by less than 15%, making it more suitable for spaceborne on-orbit applications than traditional fault-tolerant design approaches. Full article

This study presents an adaptation of the YOLOv4 deep learning algorithm for 3D object detection, addressing a critical challenge in autonomous vehicle (AV) systems: accurate real-time perception of the surrounding environment in three dimensions. Traditional 2D detection methods, while efficient, fall short in providing the depth and spatial information necessary for safe navigation. This research modifies the YOLOv4 architecture to predict 3D bounding boxes, object depth, and orientation. Key contributions include introducing a multi-task loss function that optimizes 2D and 3D predictions and integrating sensor fusion techniques that combine RGB camera data with LIDAR point clouds for improved depth estimation. The adapted model, tested on real-world datasets, demonstrates a significant increase in 3D detection accuracy, achieving a mean average precision (mAP) of 85%, intersection over union (IoU) of 78%, and near real-time performance at 93–97% for detecting vehicles and 75–91% for detecting people. This approach balances high detection accuracy and real-time processing, making it highly suitable for AV applications. This study advances the field by showing how an efficient 2D detector can be extended to meet the complex demands of 3D object detection in real-world driving scenarios without sacrificing computational efficiency. Full article

Multi-FPGA systems can form larger and more powerful computing units through high-speed interconnects between chips, and are beginning to be widely used by various computing service providers, especially in edge computing. However, the new computing architecture brings new challenges to efficient and reliable task scheduling. In this context, we propose a resource- and reliability-aware hybrid scheduling method on Multi-FPGA systems. First, a set of models is established based on the resource/time requirements, communication overhead, and state conversion process of tasks to further analyze the constraints of system scheduling. On this basis, the large task is divided into subtasks based on the data dependency matrix, and the Maintenance Multiple Sequence (MMS) algorithm is used to generate execution sequences for each subtask to the Multi-FPGA systems to fully exploit resources and ensure reliable operation. Compared with state-of-the-art scheduling methods, the proposed method can achieve an average increase in resource utilization of 7%; in terms of reliability, it achieves good execution gains, with an average task completion rate of 98.3% and a mean time to failure of 15.7 years. Full article

An intelligent transportation system (ITS) offers commercial and personal movement through the smart city (SC) communication paradigms with hassle-free information sharing. ITS designs and architectures have improved via information and communication technologies in recent years. The information shared through the communication medium in SCs is exposed to adversary risk, resulting in privacy issues. Privacy issues impact the contingent mobility and localization of the ITS path. This paper introduces a novel resilient privacy preserving (RPP) method through presumed secrecy (PS) to provide a robust privacy measure. The privacy of the progressive communication sessions is preserved based on the previous security depletion levels. The interruptions in traffic data-related communication sessions are recurrently identified, and re-handoffs are recommended with dodged transfer learning. The empirical results indicate a 25% reduction in computational overhead and a 30% enhancement in privacy protection over conventional methods, demonstrating the model’s efficacy in secure ITS communication. Compared with existing methods, the proposed approach decreases security depletion rates by 15% across varying traffic densities, underscoring ITS resilience in high-interaction scenarios. Full article

Delivering persistent values in a dynamic environment is a challenging but imperative capability for a system-of-systems (SoS). Practitioners in the SoS and defense domains are exploring the benefits of the operational-level reconfiguration strategies via new operational concepts such as mosaic warfare. However, an architecture design that allows reconfiguration is also a crucial task, but has not yet received adequate attention, not to mention accounting for the mutual impact between architecture design alternatives and reconfiguration options. Therefore, this paper proposes an integrated method that can select the architecture with a specific inherent structure in the design phase that supports dynamic reconfiguration during the operational phase. This method firstly builds a structural framework that connects architecture design and reconfiguration, and identifies the enablers for SoS architecture reconfiguration. After developing an SoS effectiveness evaluator, the method constructs an integrated multi-objective formulation for the initial architecture selection and reconfiguration process, and provides a solution algorithm based on a fast non-dominated sorting genetic algorithm. An application to an air and missile defense SoS illustrates the effectiveness of the proposed method. The generated Pareto optimal set of solutions that have non-dominated recoverability and survivability provide useful decision support for SoS composition and initial architecture configuration, based upon which an SoS can also respond effectively to disruptions by computing the reconfiguration decisions. Full article

Companies have tried to innovate in their training processes to increase their productivity indicators, reduce equipment maintenance costs, and improve the work environment. The use of Augmented Reality (AR) has been one of the implemented strategies to upgrade training processes, since it optimizes, through User Interface (UI) Design, experiences designed for users (UX) that are focused on education and training contexts. This research describes the definition and implementation of an IT architecture based on the ISO/IEC/IEEE 42010 standard using the Zachman and Kruchten frameworks. The methodological proposal presents an architecture seen from a business perspective, taking into account the strategic and technological components of the organization under a strategic alignment approach. The result is a six-layer architecture: The Government Strategy Layer (1) that accounts for the strategic component; the Business Layer (2) that presents the business management perspective; the Information Layer (4) that defines the metrics system: efficiency through task time, effectiveness through tasks completed, and satisfaction with overall satisfaction. In the Data Layer (4), the data collected with the metrics are structured in an industrial scenario with a cylinder turning process on a Winston Lathe. The experiment was carried out with two groups of 272 participants. In the Systems and Applications Layer (5), two applications were designed: a web client and a mobile application with augmented reality, and finally, the Networks and Infrastructure Layer (6), which delivers the two functional applications. The architecture validation was carried out using the mobile application. The analysis of the results showed a significance value of less than 0.001 in the three indicators: efficiency, effectiveness, and satisfaction in the Levene test and Student’s t-test. To corroborate the results, a test of equality of means with the Mann–Whitney U was carried out, showing that the three indicators presented significantly different values in the two experimental groups of this study. Thus, the group trained with the application obtained better results in the three indicators. The proposed architecture is adaptable to other training contexts. Information, data, and systems and application layers allowed for the exchange of training processes so that the augmented reality application is updated according to the new requirements. Full article

This paper presents a comprehensive study of the application of AI-driven inpainting techniques to the restoration of historical photographs of the Czech city Most, with a focus on restoration and reconstructing the lost architectural heritage. The project combines state-of-the-art methods, including generative adversarial networks (GANs), patch-based inpainting, and manual retouching, to restore and enhance severely degraded images. The reconstructed/restored photographs of the city Most offer an invaluable visual representation of a city that was largely destroyed for industrial purposes in the 20th century. Through a series of blind and informed user tests, we assess the subjective quality of the restored images and examine how knowledge of edited areas influences user perception. Additionally, this study addresses the technical challenges of inpainting, including computational demands, interpretability, and bias in AI models. Ethical considerations, particularly regarding historical authenticity and speculative reconstruction, are also discussed. The findings demonstrate that AI techniques can significantly contribute to the preservation of cultural heritage, but must be applied with careful oversight to maintain transparency and cultural integrity. Future work will focus on improving the interpretability and efficiency of these methods, while ensuring that reconstructions remain historically and culturally sensitive. Full article

Due to its non-contact characteristics, remote photoplethysmography (rPPG) has attracted widespread attention in recent years, and has been widely applied for remote physiological measurements. However, most of the existing rPPG models are unable to estimate multiple physiological signals simultaneously, and the performance of the limited available multi-task models is also restricted due to their single-model architectures. To address the above problems, this study proposes MultiPhys, adopting a heterogeneous network fusion approach for its development. Specifically, a Convolutional Neural Network (CNN) is used to quickly extract local features in the early stage, a transformer captures global context and long-distance dependencies, and Mamba is used to compensate for the transformer’s deficiencies, reducing the computational complexity and improving the accuracy of the model. Additionally, a gate is utilized for feature selection, which classifies the features of different physiological indicators. Finally, physiological indicators are estimated after passing features to each task-related head. Experiments on three datasets show that MultiPhys has superior performance in handling multiple tasks. The results of cross-dataset and hyper-parameter sensitivity tests also verify its generalization ability and robustness, respectively. MultiPhys can be considered as an effective solution for remote physiological estimation, thus promoting the development of this field. Full article

Accurate and timely air quality forecasting is crucial for mitigating pollution-related hazards and protecting public health. Recently, there has been a growing interest in integrating visual data for air quality prediction. However, some limitations remain in existing literature, such as their focus on coarse-grained classification, single-moment estimation, or reliance on indirect and unintuitive information from visual images. Here we present a dual-channel deep learning model, integrating surveillance images and multi-source numerical data for air quality forecasting. Our model, which combines a single-channel hybrid network consisting of VGG16 and LSTM (named VGG16-LSTM) with a single-channel Long Short-Term Memory (LSTM) network, efficiently captures detailed spatiotemporal features from surveillance image sequences and temporal features from atmospheric, meteorological, and temporal data, enabling accurate time-series forecasting of PM_2.5 and PM₁₀ concentrations. Experiments conducted on the 2021 Shanghai dataset demonstrate that the proposed model significantly outperforms traditional machine learning methods in terms of accuracy and robustness for time-series forecasting, achieving R² values of 0.9459 and 0.9045 and RMSE values of 4.79 μg/m³ and 11.51 μg/m³ for PM_2.5 and PM₁₀, respectively. Furthermore, validation results on the datasets from two stations in Kaohsiung, Taiwan, with average R² values of 0.9728 and 0.9365 and average RMSE values of 1.89 μg/m³ and 5.69 μg/m³ for PM_2.5 and PM₁₀ using a pretrain–finetune training strategy, confirm the model’s adaptability across diverse geographical contexts. These findings highlight the potential of integrating surveillance images to enhance air quality prediction, offering an effective supplement to ground-level environmental monitoring. Future work will focus on expanding datasets and optimizing network architectures to further improve forecasting accuracy and computational efficiency, enhancing the model’s scalability for broader regional air quality management. Full article

Reconfigurable processor-based acceleration of deep convolutional neural network (DCNN) algorithms has emerged as a widely adopted technique, with particular attention on sparse neural network acceleration as an active research area. However, many computing devices that claim high computational power still struggle to execute neural network algorithms with optimal efficiency, low latency, and minimal power consumption. Consequently, there remains significant potential for further exploration into improving the efficiency, latency, and power consumption of neural network accelerators across diverse computational scenarios. This paper investigates three key techniques for hardware acceleration of sparse neural networks. The main contributions are as follows: (1) Most neural network inference tasks are typically executed on general-purpose computing devices, which often fail to deliver high energy efficiency and are not well-suited for accelerating sparse convolutional models. In this work, we propose a specialized computational circuit for the convolutional operations of sparse neural networks. This circuit is designed to detect and eliminate the computational effort associated with zero values in the sparse convolutional kernels, thereby enhancing energy efficiency. (2) The data access patterns in convolutional neural networks introduce significant pressure on the high-latency off-chip memory access process. Due to issues such as data discontinuity, the data reading unit often fails to fully exploit the available bandwidth during off-chip read and write operations. In this paper, we analyze bandwidth utilization in the context of convolutional accelerator data handling and propose a strategy to improve off-chip access efficiency. Specifically, we leverage a compiler optimization plugin developed for Vitis HLS, which automatically identifies and optimizes on-chip bandwidth utilization. (3) In coefficient-based accelerators, the synchronous operation of individual computational units can significantly hinder efficiency. Previous approaches have achieved asynchronous convolution by designing separate memory units for each computational unit; however, this method consumes a substantial amount of on-chip memory resources. To address this issue, we propose a shared feature map cache design for asynchronous convolution in the accelerators presented in this paper. This design resolves address access conflicts when multiple computational units concurrently access a set of caches by utilizing a hash-based address indexing algorithm. Moreover, the shared cache architecture reduces data redundancy and conserves on-chip resources. Using the optimized accelerator, we successfully executed ResNet50 inference on an Intel Arria 10 1150GX FPGA, achieving a throughput of 497 GOPS, or an equivalent computational power of 1579 GOPS, with a power consumption of only 22 watts. Full article

This research concerns the exterior lighting of historic buildings and cultural heritage monuments. Its objective is to organize a methodology for the study of facades, to record the individual or grouped morphological and decorative elements of the facades, and to organize the steps to achieve a presentation of different ways of lighting these elements. This presentation is made by an experimental digital lighting simulation, leading the researcher to discover the relationship between light and the architectural element being illuminated. Finally, the results of the simulations are evaluated by experts in the field of lighting, who attest to the emotions generated by the observation of the different lighting scenarios, while an attempt is then made to synthesize these results on an entire building facade, to determine whether this synthesis of the individual lighting effects is practicable. The analysis of the results reveals the trends in each lighting scenario, leading to a variety of emotions, whether they arise from a specific morphological element or from the entire facade. Full article

In various practical noise control scenarios, such as duct noise mitigation, industrial machinery, architectural acoustics, and underwater applications, it is essential to develop noise absorbers that deliver effective low-frequency attenuation while maintaining compact dimensions. To achieve low-frequency absorption within a limited spatial volume, this study proposes an embedded Helmholtz resonator featuring a roughened neck and establishes a numerical computational model that incorporates thermos viscous effects. A quantitative investigation is conducted on three types of embedded rough-neck geometries (rectangular-grooved, triangular-grooved, and undulated) to elucidate their acoustic performance, with particular attention to differences in acoustic transmission loss and acoustic impedance characteristics. In response to the practical demand for even lower-frequency attenuation, this work further focuses on optimizing the structural parameters of an embedded rectangular-grooved Helmholtz resonator (ERHR). A back-propagation (BP) neural network models and predicts how structural parameters impact the acoustic transmission coefficient, elucidating the effects of geometric variations. Moreover, by coupling the BP network with the Golden Jackal Optimization (GJO) algorithm, a BP-GJO optimization model is developed to refine the structural parameters. The findings reveal that the proposed method significantly improves resonator spatial utilization at a specific noise frequency while preserving acoustic transmission loss performance. This work thereby provides a promising strategy for designing low-frequency, compact Helmholtz resonators suitable for a wide range of noise control applications. Full article

We conducted an in-depth exploration of the use of different machine learning (ML) for regression algorithms, including Linear, Ridge, LASSO, Bayesian Ridge, Decision Tree, and a variety of Deep Neural Network (DNN) architectures, to estimate the slope of the high-frequency feature (HFF), a prominent emergent feature found in the gravitational wave (GW) signals of core collapse supernovae (CCSN). We created a data set of CCSN GW signals generated by an analytical model that mimics the characteristics of the signals obtained from numerical simulations, particularly the HFF. This enabled us to simulate a wide range of HFF slope values and analyze their properties. We opted to employ ML for regression techniques, particularly a supervised learning approach, to analyze the data set due to the parameter chosen for estimating the slope of the HFF. This type of architecture is ideal for this purpose as it can detect the connections between input and output data. In addition, it is suitable for handling high-dimensional input data and produces efficient results with low computational cost. We evaluated the efficiency and performance of the ML algorithms using a set of metrics to measure their ability to accurately predict the HFF slope within the data set. The results showed that a DNN algorithm for regression exhibits the highest accuracy in estimating the slope of the HFF. Full article