Search Results (1,408)

As a well-known World Cultural Heritage Site, the Historic Centre of Macao’s historical buildings possess a wealth of decorative patterns. These patterns contain cultural esthetics, geographical environment, cultural traditions, and other elements from specific historical periods, deeply reflecting the evolution of religious rituals and political and economic systems throughout history. Through long-term research, this article constructs a dataset of 11,807 images of local decorative patterns of historical buildings in Macau, and proposes a fine-grained image classification method using the ConvNeXt-L model. The ConvNeXt-L model is an efficient convolutional neural network that has demonstrated excellent performance in image classification tasks in fields such as medicine and architecture. Its outstanding advantages lie in limited training samples, diverse image features, and complex scenes. The most typical advantage of this model is its structural integration of key design concepts from a Transformer, which significantly enhances the feature extraction and generalization ability of samples. In response to the objective reality that the decorative patterns of historical buildings in Macau have rich levels of detail and a limited number of functional building categories, ConvNeXt-L maximizes its ability to recognize and classify patterns while ensuring computational efficiency. This provides a more ideal technical path for the classification of small-sample complex images. This article constructs a deep learning system based on the PyTorch 1.11 framework and compares ResNet50, EfficientNet-B7, ViT-B/16, Swin-B, RegNet-Y-16GF, and ConvNeXt series models. The results indicate a positive correlation between model performance and structural complexity, with ConvNeXt-L being the most ideal in terms of accuracy in decorative pattern classification, due to its fusion of convolution and attention mechanisms. This study not only provides a multidimensional exploration for the protection and revitalization of Macao’s historical and cultural heritage and enriches theoretical support and practical foundations but also provides new research paths and methodological support for artificial intelligence technology to assist in the planning and decision-making of historical urban areas. Full article

Software-Defined Wide Area Networks (SD-WAN) have emerged as a rapidly evolving technology designed to meet the growing demand for flexible, secure, and scalable network infrastructures. This paper provides a review of SD-WAN techniques, focusing on their principles of operation, mechanisms, and evolution, with particular attention to applications in resource-constrained environments such as mobile, satellite, and radio networks. The analysis highlights key architectural elements, including security mechanisms, monitoring methods and metrics, and management protocols. A classification of both commercial (e.g., Cisco SD-WAN, Fortinet Secure SD-WAN, VMware SD-WAN, Palo Alto Prisma SD-WAN, HPE Aruba EdgeConnect) and research-based solutions is presented. The overview covers overlay protocols such as Overlay Management Protocol (OMP), Dynamic Multipath Optimization (DMPO), App-ID, OpenFlow, and NETCONF, as well as tunneling mechanisms such as IPsec and WireGuard. The discussion further covers control plane architectures (centralized, distributed, and hybrid) and network monitoring methods, including latency, jitter, and packet loss measurement. The growing importance of Artificial Intelligence (AI) in optimizing path selection and improving threat detection in SD-WAN environments, especially in resource-constrained networks, is emphasized. Analysis of solutions indicates that SD-WAN improves performance, reduces latency, and lowers operating costs compared to traditional WAN architectures. The paper concludes with guidelines and recommendations for using SD-WAN in resource-constrained environments. Full article

The analysis of dynamic remote sensing scenes from unmanned aerial vehicles (UAVs) is shifting from reactive processing to proactive, predictive intelligence. Central to this evolution is video prediction—forecasting future imagery from past observations—which enables critical remote sensing applications like persistent environmental monitoring, occlusion-robust object tracking, and infrastructure anomaly detection under challenging aerial conditions. Yet, a systematic review of video prediction models tailored for the unique constraints of aerial remote sensing has been lacking. Existing taxonomies often obscure key design choices, especially for emerging operators like state-space models (SSMs). We address this gap by proposing a unified, multi-dimensional taxonomy with three orthogonal axes: (i) operator architecture; (ii) generative nature; and (iii) training/inference regime. Through this lens, we analyze recent methods, clarifying their trade-offs for deployment on UAV platforms that demand processing of high-resolution, long-horizon video streams under tight resource constraints. Our review assesses the utility of these models for key applications like proactive infrastructure inspection and wildlife tracking. We then identify open problems—from the scarcity of annotated aerial video data to evaluation beyond pixel-level metrics—and chart future directions. We highlight a convergence toward scalable dynamic world models for geospatial intelligence, which leverage physics-informed learning, multimodal fusion, and action-conditioning, powered by efficient operators like SSMs. Full article

This paper introduces the concept of evolving actor–network ontologies (EANO) as a new paradigm for cultural digital twins. Building on actor–network theory, EANO reframes ontologies from static representations into reflexive, dynamic structures in which semantic interpretations are continuously negotiated among heterogeneous actors. We propose a five-layer architecture that operationalizes this principle, embedding reflexivity, actor salience, and systemic parameters such as resistance and volatility directly into the ontological model. To illustrate this approach, we present minimal simulations that demonstrate how different actor constellations and systemic conditions lead to distinct patterns of semantic evolution, ranging from expert erosion to contested equilibria and balanced coexistence. Rather than serving as predictive models, these simulations exemplify how EANO captures semantic plurality and contestation within a transparent and interpretable framework. The contribution of this work is thus twofold: it provides a conceptual foundation for evolving ontologies in digital heritage and a lightweight demonstration of how such models can be instantiated and explored computationally. Full article

The indigenous community of Quisquinchir, in Saraguro (Loja, Ecuador), is facing a process of transformation of the rural Andean landscape associated with internal and external migration, as well as the influence of foreign architectural models. The new buildings symbolize, in the collective imagination, modernity and progress; however, they are alien to the natural environment characterized by the practice of agricultural and livestock activities. Although previous studies have described the loss of Andean vernacular architecture, its recent evolution in clear typologies has not been systematized. The objective of this study is to assess the current state of traditional dwellings and understand how migration reconfigures the landscape, collective memory, building traditions, and cultural identity of their inhabitants. Based on direct observation, photographic and stratigraphic analysis, and secondary sources, five typologies were identified: traditional one-story, traditional two-story, hybrid one-story, hybrid two-story, and eclectic. This classification indicates the replacement of earthen walls with cement blocks in 37% of the dwellings and of tile roofs with zinc roofs in 29%. However, 35% of the houses retain their traditional morphology and materials. These results and their classification are fundamental contributions to the design of local public policies that generate adequate interventions respectful of the environment. Full article

Battery Electric Vehicles (BEVs) technology is rapidly emerging as the cornerstone of sustainable transportation, driven by advancements in battery technology, power electronics, and modern drivetrains. This paper presents a comprehensive review of current and next-generation BEV powertrain architectures, focusing on five key subsystems: battery energy storage system, electric propulsion motors, energy management systems, power electronic converters, and charging infrastructure. The review traces the evolution of battery technology from conventional lithium-ion to solid-state chemistries and highlights the critical role of battery management systems in ensuring optimal state of charge, health, and safety. Recent innovations by leading automakers are examined, showcasing advancements in cell formats, motor designs, and thermal management for enhanced range and performance. The role of power electronics and the integration of AI-driven strategies for vehicle control and vehicle-to-grid (V2G) are analyzed. Finally, the paper identifies ongoing research gaps in system integration, standardization, and advanced BMS solutions. This review provides a comprehensive roadmap for innovation, aiming to guide researchers and industry stakeholders in accelerating the adoption and sustainable advancement of BEV technologies. Full article

The evolution of Industrial Internet of Things (IIoT) systems demands flexible and intelligent architectures capable of addressing low-latency requirements, real-time analytics, and adaptive resource management. In this context, softwarized edge computing emerges as a key enabler, supporting advanced IoT deployments through programmable infrastructures, distributed intelligence, and seamless integration with cloud environments. This paper presents an extended and publicly available proof of concept (PoC) for a softwarized, data-driven architecture designed to operate across the cloud/edge/IoT continuum. The proposed architecture incorporates containerized microservices, open standards, and ML-based inference services to enable runtime decision-making and on-the-fly network reconfiguration based on real-time telemetry from IIoT nodes. Unlike traditional solutions, our approach leverages a modular control plane capable of triggering dynamic adaptations in the system through RESTful communication with a cloud-hosted inference engine, thus enhancing responsiveness and autonomy. We evaluate the system in representative IIoT scenarios involving multi-agent collaboration, showcasing its ability to process data at the edge, minimize latency, and support real-time decision-making. This work contributes to the ongoing efforts toward building advanced IoT ecosystems by bridging conceptual designs and practical implementations, offering a robust foundation for future research and deployment in intelligent, software-defined industrial environments. Full article

Drawing upon the works of Kazadi wa Mukuma, Gerhard Kubik, Carlos de Lima, Vivian Gotheim, Wilson Nogueira, Temitope Fagunwa, and Alaba Simpson, this study traced the evolution of Bumba-Meu-Boi from its regional origins in Maranhao, Brazil, to its adaptation in Lagos, Nigeria, as an Afro-Brazilian returnee festival within the context of Lagos carnival. Beyond serving as a crucible for the historical return of repatriated Africans from Brazil following abolition of slavery in Brazil, the study also documents how the Afro-Brazilian community has been fully integrated into the Nigerian society. Through the formation of a thriving Brazilian Descendants Association, the Brazilian community has been able to sustain their Afro-Brazilian heritage through social events and community impact by preserving Brazilian architecture, culinary knowledge, festivals, teaching of Portuguese language, and the celebration of their Afro-Brazilian genealogical past. Full article

This review paper analyzes the recent applications of computational mechanics methods in combination with machine learning (ML) and neural network (NN) techniques, as found in the literature published between 2015 and 2024. We present how ML and NNs are enhancing traditional computational methods, such as the finite element method, enabling the solution of complex problems in material modeling, surrogate modeling, inverse analysis, and uncertainty quantification. We categorize current research by considering the specific computational mechanics tasks and the employed ML/NN architectures. Furthermore, we discuss the current challenges, development opportunities, and future directions of this dynamically evolving interdisciplinary field, highlighting the potential of data-driven approaches to transform the modeling and simulation of mechanical systems. The review has been updated to include pivotal publications from 2025, reflecting the rapid evolution of the field in multiscale modeling, data-driven mechanics, and physics-informed/operator learning. Accordingly, the timespan is now 2015–2025, with a focused inclusion of high-impact contributions from 2024 to 2025. Full article

Background: Electrochemical impedance spectroscopy (EIS) is indispensable for disentangling charge-transfer, capacitive, and diffusive phenomena, yet reproducible parameter estimation and objective model selection remain unsettled. Methods: We derive closed-form impedances and analytical Jacobians for seven equivalent-circuit models (Randles, constant-phase element (CPE), and Warburg impedance (ZW) variants), enforce physical bounds, and fit synthetic spectra with 2.5% and 5.0% Gaussian noise using hybrid optimization (Differential Evolution (DE) → Levenberg–Marquardt (LM)). Uncertainty is quantified via non-parametric bootstrap; parsimony is assessed with root-mean-square error (RMSE), Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC); physical consistency is checked by Kramers–Kronig (KK) diagnostics. Results: Solution resistance ($R_s$) and charge-transfer resistance ($R_{ct}$) are consistently identifiable across noise levels. CPE parameters $(Q,n)$ and diffusion amplitude $\left(\sigma \right)$ exhibit expected collinearity unless the frequency window excites both processes. Randles suffices for ideal interfaces; Randles+CPE lowers AIC when non-ideality and/or higher noise dominate; adding Warburg reproduces the ${45}^{\circ }$ tail and improves likelihood when diffusion is present. The $(R_{ct}+Z_W)\Vert \mathrm{CPE}$ architecture offers the best trade-off when heterogeneity and diffusion coexist. Conclusions: The framework unifies analytical derivations, hybrid optimization, and rigorous statistics to deliver traceable, reproducible EIS analysis and clear applicability domains, reducing subjective model choice. All code, data, and settings are released to enable exact reproduction. Full article

The primary aim of this study is to develop an effective decision-support system for managing crises related to the release of hazardous airborne substances. Such incidents, which can arise from industrial accidents or intentional releases, necessitate the rapid identification of contaminant sources to enable timely response measures. This work focuses on a novel approach that integrates a modified Sandpile model with advection and employs the (1 + 1)-Evolution Strategy to solve the inverse problem of source localization. The initial section of this paper reviews existing methods for simulating atmospheric dispersion and reconstructing source locations. In the following sections, we describe the architecture of the proposed system, the modeling assumptions, and the experimental framework. A key feature of the method presented here is its reliance solely on concentration measurements obtained from a distributed network of sensors, eliminating the need for prior knowledge of the source location, release time, or emission strength. The system was validated through a two-stage process using synthetic data generated by a Gaussian dispersion model. Preliminary experiments were conducted to support model calibration and refinement, followed by formal tests to evaluate localization accuracy and robustness. Each test case was completed in under 20 min on a standard laptop, demonstrating the algorithm’s high computational efficiency. The results confirm that the proposed (1 + 1)-ES Sandpile model can effectively reconstruct source parameters, staying within the resolution limits of the sensor grid. The system’s speed, simplicity, and reliance exclusively on sensor data make it a promising solution for real-time environmental monitoring and emergency response applications. Full article

Urban roof segmentation plays a pivotal role in applications such as urban planning, infrastructure management, and renewable energy deployment. This study explores the evolution of deep learning techniques from traditional Convolutional Neural Networks (CNNs) to cutting-edge transformer-based models in the context of roof segmentation from satellite imagery. We highlight the limitations of conventional methods when applied to urban environments, including resolution constraints and the complexity of roof structures. To address these challenges, we evaluate two advanced deep learning models, Mask R-CNN and MaskFormer, which have shown significant promise in accurately segmenting roofs, even in dense urban settings with diverse roof geometries. These models, especially the one based on transformers, offer improved segmentation accuracy by capturing both global and local image features, enhancing their performance in tasks where fine detail and contextual awareness are critical. A case study on Ben Guerir City in Morocco, an urban area experiencing rapid development, serves as the foundation for testing these models. Using high-resolution satellite imagery, the segmentation results offer a deeper understanding of the accuracy and effectiveness of these models, particularly in optimizing urban planning and renewable energy assessments. Quantitative metrics such as Intersection over Union (IoU), precision, recall, and F1-score are used to benchmark model performance. Mask R-CNN achieved a mean IoU of 74.6%, precision of 81.3%, recall of 78.9%, and F1-score of 80.1%, while MaskFormer reached a mean IoU of 79.8%, precision of 85.6%, recall of 82.7%, and F1-score of 84.1% (pixel-level, micro-averaged at IoU = 0.50 on the held-out test set), highlighting the transformative potential of transformer-based architectures for scalable and precise urban imaging. The study also outlines future work in 3D modeling and height estimation, positioning these advancements as critical tools for sustainable urban development. Full article

Dermatopathology has rapidly advanced due to the implementation of deep learning models and artificial intelligence (AI). From convolutional neural networks (CNNs) to transformer-based foundation models, these systems are now capable of accurate whole-slide analysis and multimodal integration. This review synthesizes the most recent advents of deep-learning architecture and synthesizes its evolution from first-generation CNNs to hybrid CNN-transformer systems to large-scale foundational models such as Paige’s PanDerm AI and Virchow. Herein, we examine performance benchmarks from real-world deployments of major dermatopathology deep learning models (DermAI, PathAssist Derm), as well as emerging next-generation models still under research and development. We assess barriers to clinical workflow adoption such as dataset bias, AI interpretability, and government regulation. Further, we discuss potential future research directions and emphasize the need for diverse, prospectively curated datasets, explainability frameworks for trust in AI, and rigorous compliance to Good Machine-Learning-Practice (GMLP) to achieve safe and scalable deep learning dermatopathology models that can fully integrate into clinical workflows. Full article

Automatic speech recognition (ASR) has advanced rapidly, evolving from early template-matching systems to modern deep learning frameworks. This review systematically traces ASR’s technological evolution across four phases: the template-based era, statistical modeling approaches, the deep learning revolution, and the emergence of large-scale models under diverse learning paradigms. We analyze core technologies such as hidden Markov models (HMMs), Gaussian mixture models (GMMs), recurrent neural networks (RNNs), and recent architectures including Transformer-based models and Wav2Vec 2.0. Beyond algorithmic development, we examine how ASR integrates into intelligent information systems, analyzing real-world applications in healthcare, education, smart homes, enterprise systems, and automotive domains with attention to deployment considerations and system design. We also address persistent challenges—noise robustness, low-resource adaptation, and deployment efficiency—while exploring emerging solutions such as multimodal fusion, privacy-preserving modeling, and lightweight architectures. Finally, we outline future research directions to guide the development of robust, scalable, and intelligent ASR systems for complex, evolving environments. Full article

This paper proposes a novel direct-drive rotary actuator based on a modular five-phase outer-rotor flux-switching permanent magnet (FSPM) machine to overcome the limitations of conventional actuators with gear reducers, such as mechanical complexity and low reliability. The research focused on a synergistic design of a lightweight, high-torque-density motor and a precise control strategy. The methodology involved a structured topology evolution to create a modular stator architecture, followed by finite element analysis-based electromagnetic optimization. To achieve precision control, a multi-vector model predictive current control (MPCC) scheme was developed. This optimization process contributed to a significant performance improvement, increasing the average torque to 13.33 Nm, reducing torque ripple from 9.81% to 2.36% and obtaining a maximum position error under 1 mil. The key result was experimentally validated using an 8 kg inertial load, confirming the actuator’s feasibility for industrial deployment. Full article