Saved Queries

Perceived landscape quality in historical and cultural districts is crucial for reconciling cultural heritage preservation with urban renewal. However, limited attention has been paid to whether street-level visual elements influence public esthetic perception in a nonlinear manner, especially in heritage-sensitive urban environments. Against this backdrop, this study explores the nonlinear effects of natural, artificial, and interfering visual elements on perceived landscape quality in historical and cultural districts. Five districts in Wuhan, China, were selected. Street view images were processed with a U-Net–based semantic segmentation model to extract pixel-level visual elements, and public scenic beauty ratings were collected through an image-based questionnaire survey. The analyses reveal nonlinear perception patterns. The results show that the relationships between visual elements and perceived beauty are nonlinear and heterogeneous. Natural elements have the strongest positive influence on perceived landscape quality, artificial elements require careful density control, and interfering elements are consistently negative contributors. By quantifying these nonlinear mechanisms, this study suggests that esthetic responses in historical districts may depend on threshold-like combinations of visual elements and may offer a useful reference for heritage-sensitive urban renewal and streetscape design. Full article

(This article belongs to the Section Architectural Design, Urban Science, and Real Estate)

28 pages, 2001 KB

Open AccessArticle

NIABIAuth: A Non-Interactive Attribute Binding Identity Authentication Protocol for Internet of Things Terminals

by Zilong Han, Xinge Li, Chaoqun Kang, Haowen Sun and Yali Gao

Information 2025, 16(12), 1040; https://doi.org/10.3390/info16121040 (registering DOI) - 28 Nov 2025

Abstract

The large-scale deployment of distributed devices in the Internet of Things (IoT) brings urgent demands for secure, scalable, and lightweight identity authentication. For example, virtual power plants integrate numerous heterogeneous energy terminals to support grid dispatch and market operations, while posing challenges such as real-time access, resource constraints, and identity privacy protection. To address these challenges, this paper proposes NIABIAuth, a non-interactive attribute binding identity authentication protocol for IoT terminals. NIABIAuth supports dynamic challenge computation and binds cryptographic identity proofs with terminal attributes, enabling fine-grained and privacy preserving access control. By storing identity credentials and verification records on the chain, this protocol ensures traceability and tamper resistance. Experiments demonstrate that NIABIAuth maintains low authentication latency and is consistent throughput, even under constrained conditions. Compared with baseline methods, NIABIAuth achieves substantial reductions in communication and computation cost. The proposed NIABIAuth was formally verified using the AVISPA tool, which proved that it could resist common attacks, including replay attacks, man-in-the-middle attacks, etc. A large number of simulation experiments have demonstrated that the proposed protocol can provide real-time identity authentication for Internet of Things terminals. Full article

14 pages, 1030 KB

Open AccessArticle

No Correlation Between Proteinuria and Renal Function in Patients with Unresectable Hepatocellular Carcinoma Treated with Atezolizumab Plus Bevacizumab: ARISE Study

by Kazuomi Ueshima, Naoshi Nishida, Satoru Hagiwara, Yasunori Minami, Hiroshi Ida, Masahiro Takita, Hirokazu Chishina, Masahiro Morita, Tomoko Aoki, Tetsutaro Hamano, Ryosuke Take, Chizuko Watanabe, Kohsuke Asoh, Ai Tanaka and Masatoshi Kudo

Cancers 2025, 17(23), 3826; https://doi.org/10.3390/cancers17233826 (registering DOI) - 28 Nov 2025

Abstract

Background: Atezolizumab plus bevacizumab (Atezo + Bev) is the standard of care for treatment-naïve patients with unresectable hepatocellular carcinoma (uHCC). Proteinuria is a treatment-emergent adverse event that often leads to Bev interruption. However, the relationship between Bev-related proteinuria and renal dysfunction is unclear. We retrospectively investigated the impact of proteinuria after starting Atezo + Bev on renal function in patients with uHCC. Methods: We performed a single-arm retrospective study of patients with uHCC treated with Atezo + Bev between 25 September 2020 and 31 May 2022, at Kindai University Hospital, Japan. The impact of proteinuria on renal function during Atezo + Bev treatment was analyzed in terms of the correlation between changes in urine protein creatinine ratio (UPCR) and estimated glomerular filtration rate (eGFR) relative to baseline. Results: We analyzed data from 100 patients (median age 74 years; range 41–89; 75% male). During Atezo + Bev treatment, the median (interquartile range) maximum increase from baseline in UPCR was 0.39 (0.08 to 2.05) and the median maximum decline from baseline in eGFR was −7.5 (−20.5 to −3.0) mL/min/1.73 m². The Pearson and Spearman correlation coefficients (95% confidence intervals) between these variables were −0.16 (−0.34 to 0.04) and −0.13 (−0.32 to 0.07), respectively. Conclusions: We found no correlation between the changes in UPCR and eGFR during Atezo + Bev treatment. Bev interruption criteria are based on the degree of proteinuria; however, our results suggest that proteinuria does not necessarily impair renal function. Physicians should consider the risk–benefit profile when deciding whether to discontinue Bev in patients who develop proteinuria during Atezo + Bev treatment. Full article

(This article belongs to the Special Issue The Biology of Hepatocellular Carcinoma: Implications for Genomic and Immune Therapies (2nd Edition))

62 pages, 1053 KB

Open AccessReview

Challenge of Corneal Ulcer Healing: A Novel Conceptual Framework, the “Triad” of Corneal Ulcer Healing/Corneal Neovascularization/Intraocular Pressure, and Avascular Tendon Healing, for Evaluation of Corneal Ulcer Therapy, Therapy of Neovascularization, Glaucoma Therapy, and Pentadecapeptide BPC 157 Efficacy

by Sanja Masnec, Antonio Kokot, Tamara Kralj, Mirna Zlatar, Kristina Loncaric, Marko Sablic, Miro Kalauz, Iva Beslic, Katarina Oroz, Bozana Mrvelj, Lidija Beketic Oreskovic, Ivana Oreskovic, Sanja Strbe, Borna Staresinic, Goran Slivsek, Alenka Boban Blagaic, Sven Seiwerth, Anita Skrtic and Predrag Sikiric

Pharmaceuticals 2025, 18(12), 1822; https://doi.org/10.3390/ph18121822 (registering DOI) - 28 Nov 2025

Abstract

To better address the challenge of corneal ulcer healing, with already available standard agents, and those recently introduced, such as stable gastric pentadecapeptide BPC 157, we introduced a novel conceptual framework—the “triad” of corneal ulcer healing↔corneal neovascularization↔intraocular pressure—and extended it to avascular tissues such as tendon. Within this framework, cytoprotection serves as the unifying principle, underscoring that therapeutic effects are not isolated but interconnected. Preclinical studies with BPC 157 therapy, as a cytoprotection agent, illustrate this integration. BPC 157 rapidly normalizes elevated intraocular pressure in glaucomatous rats, preserves retinal integrity, restores pupil function, maintains corneal transparency during ulcer or abrasion healing, and counteracts both corneal neovascularization and dry eye. In parallel, its consistent efficacy in tendon injury models highlights a cytoprotective specificity across avascular tissues. The cornea’s “angiogenic privilege,” preserved during healing and tendon recovery together, provides strong proof of concept. Furthermore, mapping standard therapeutic agents used for corneal ulcers, neovascularization, or glaucoma onto this triad, and linking them with tendon healing, reveals both shared pathways and inconsistencies across existing drug classes. Analyzed were the ascorbate, fibronectin, hyaluronic acid, metalloproteinase inhibitors, EGF, FGF, NGF, insulin, and IGF-1 (corneal ulcer healing), the antiangiogenic agents (endostatin, PAI-1, PEDF, angiostatin, TSP-1, TSP-2, IFN-α), corticosteroids, NSAIDs, cyclosporine A, anti-VEGF drops (treatment of corneal neovascularization), and alpha 2-agonists, beta-blockers, carboanhydrase inhibitors, muscarinic agonists, Rho-kinase inhibitors, and prostaglandin analogs (glaucoma). Taken together, these findings advance cytoprotection as a unifying therapeutic paradigm, with BPC 157 emerging as its first exemplar, and encourage further translational research toward clinical application. Full article

(This article belongs to the Special Issue Pleiotropic Agents and Their Therapeutic Potential in Gastroenterology, Angiology, and Musculoskeletal Disorders)

34 pages, 2438 KB

Open AccessArticle

Multi-Task Learning for Ocean-Front Detection and Evolutionary Trend Recognition

by Qi He, Anqi Huang, Lijia Geng, Wei Zhao and Yanling Du

Remote Sens. 2025, 17(23), 3862; https://doi.org/10.3390/rs17233862 (registering DOI) - 28 Nov 2025

Abstract

Ocean fronts are central to upper-ocean dynamics and ecosystem processes, yet recognizing their evolutionary trends from satellite data remains challenging. We present a 3D U-Net-based multi-task framework that jointly performs ocean-front detection (OFD) and ocean-front evolutionary trend recognition (OFETR) from sea surface temperature gradient heatmaps. Instead of cascading OFD and OFETR in separate stages that pass OFD outputs downstream and can amplify upstream errors, the proposed model shares 3D spatiotemporal features and is trained end-to-end. We construct the Zhejiang–Fujian Coastal Front Mask (ZFCFM) and Evolutionary Trend (ZFCFET) datasets from ESA SST CCI L4 products for 2002–2021 and use them to evaluate the framework against 2D CNN baselines and traditional methods. Multi-task learning improves OFETR compared with single-task training while keeping OFD performance comparable, and the unified design reduces parameter count and daily computational cost. The model outputs daily point-level trend labels aligned with the dataset’s temporal resolution, indicating that end-to-end multi-task learning can mitigate error propagation and provide temporally resolved estimates. Full article

(This article belongs to the Special Issue Computer Vision and Pattern Recognition for the Analysis of 2D/3D Remote Sensing Data in Geoscience (Second Edition))

18 pages, 1492 KB

Open AccessReview

Mpox in People Living with HIV: Clinical Challenges, Preventive Strategies and Public Health Implications

by Rita Cordeiro, João Caria, Daniel Sobral and Diana Póvoas

Viruses 2025, 17(12), 1558; https://doi.org/10.3390/v17121558 (registering DOI) - 28 Nov 2025

Abstract

Monkeypox virus (MPXV) re-emerged in 2022 with a global outbreak that affected more than 100,000 individuals worldwide. People living with HIV (PLWH) accounted for a substantial proportion of cases, raising concerns about disease presentation, management, and outcomes in this population. Evidence indicates that PLWH with advanced or uncontrolled HIV infection experienced more severe mpox, with higher hospitalization rates, more complications, and longer disease courses. In contrast, individuals with well-controlled HIV generally had outcomes similar to those without HIV. Access to timely diagnosis, consistent antiretroviral therapy, and availability of tecovirimat were key factors influencing prognosis. Reports also suggest bidirectional interactions between mpox and HIV pathogenesis. Immune activation and APOBEC3-related viral evolution have been proposed; however, these mechanisms remain incompletely characterized and warrant further investigation. Moreover, disparities in healthcare access and stigma compound the vulnerability of PLWH, emphasizing the need for integrated approaches. Full article

(This article belongs to the Special Issue Epidemiology of Zoonotic Virus Transmission: Pathways, Host Factors, and Public Health Strategies)

►▼ Show Figures

Graphical abstract

27 pages, 1353 KB

Open AccessArticle

Government-Led Digital Governance and the Digital Divide Among Cities: Implications for Sustainable Digital Transformation in China

by Changping Zhang, Shuai Wu, Yingying Dong and Menghan Jiang

Sustainability 2025, 17(23), 10700; https://doi.org/10.3390/su172310700 (registering DOI) - 28 Nov 2025

Abstract

Drawing on panel data from 279 prefecture-level cities in China from 2011 to 2022, this study employs the National Pilot Policy of Information Benefiting the People (NPIB) as a quasi-natural experiment to examine how government-led digital governance shapes the digital divide among cities. Using a difference-in-differences (DID) design combined with mediation and spatial analyses, the results demonstrate that the NPIB policy significantly narrowed inter-city digital disparities, with findings robust across alternative model specifications and placebo tests. Mechanism analysis shows that digital governance promotes inclusion primarily through three pathways: strengthening strategic policy orientation, enhancing technological innovation capacity, and stimulating digital market vitality. Heterogeneity analysis indicates that policy effects vary by regional development, urbanization level, and fiscal autonomy, being most pronounced in eastern cities and those with moderate urbanization and fiscal self-sufficiency. Spatial analysis reveals that while digital governance improves local inclusion, it can generate negative spillovers among neighboring cities with similar economic structures, partially offsetting aggregate gains. Overall, the findings highlight the importance of regionally differentiated strategies, cross-regional coordination, and sustained investment in digital infrastructure to promote balanced, inclusive, and sustainable digital transformation—providing practical insights for developing countries aiming to bridge structural divides and advance digital sustainability. Full article

16 pages, 576 KB

Open AccessReview

Complete Pathological Response to Neoadjuvant Treatment in Pancreatic Ductal Adenocarcinoma—Can We Achieve a Long-Term Survival? A Narrative Review

by Magdalena Gajda, Ewa Grudzińska, Łukasz Liszka, Joanna Pilch-Kowalczyk and Sławomir Mrowiec

Life 2025, 15(12), 1833; https://doi.org/10.3390/life15121833 (registering DOI) - 28 Nov 2025

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is one of the leading causes of cancer-related mortality in Europe, with a 5-year survival rate of approximately 10%. Surgical intervention is the only curative method of treatment in PDAC. However, especially in the case of patients with borderline or locally advanced cancer, neoadjuvant treatment is often administered in an attempt to downstage the tumor. Uncommonly, after neoadjuvant treatment, no viable tumor in the specimen after surgical resection is found- this is defined as a complete pathological response (pCR). Our paper presents a narrative review of this rare phenomenon and its possible association with patient’s survival. Conclusions: Achieving pCR may be associated with a significant improvement in the prognosis of patients with PDAC. However, it remains unknown why pCR is achievable in only a few patients. Further studies on large groups of patients are needed to identify the factors that increase the chance of pCR. Full article

(This article belongs to the Special Issue Pancreatic Cancer: Long Survivals)

24 pages, 3912 KB

Open AccessArticle

Single-Cell Transcriptomics of Human Acute Myocardial Infarction Reveals Oxidative Stress-Associated Cardiomyocyte Subpopulations and Candidate Predictive Signatures

by Jiashuo Hu, Ao Wang and Lan Hong

Antioxidants 2025, 14(12), 1435; https://doi.org/10.3390/antiox14121435 (registering DOI) - 28 Nov 2025

Abstract

Excessive oxidative stress drives pathological ventricular remodeling after acute myocardial infarction (AMI), yet adaptive cardiomyocyte mechanisms are poorly understood. We analyzed 64,510 human cardiomyocytes from five integrated single-cell datasets to delineate oxidative stress heterogeneity. Using quartile thresholds of a composite oxidative stress score, cells were stratified into three distinct subpopulations: high oxidative stress (HOX, score > 2.608), dynamic transient oxidative stress (DTOX), and low oxidative stress (LOX, score < 2.061). Paradoxically, HOX cells exhibited severe oxidative stress alongside significantly higher cellular plasticity than DTOX and LOX cells (p < 0.001), as confirmed by CytoTRACE and pseudotime trajectory analyses. This subpopulation demonstrated a unique “metabolic activation–immune suppression” signature and served as a central communication hub. An integrative machine-learning framework incorporating six distinct algorithms and independent cohort validation identified five core marker genes (TRIM63, ETFDH, TXNIP, CKMT2, and PDK4). These genes demonstrated stable diagnostic capability for AMI in independent validation cohorts (AUCs 0.688–0.721, all p < 0.001) and were specifically enriched in HOX cells. Our work reveals a previously unrecognized adaptive state in post-infarction cardiomyocytes, offering promising new targets for precision diagnosis and intervention. Full article

(This article belongs to the Section Aberrant Oxidation of Biomolecules)

28 pages, 4343 KB

Open AccessArticle

Investigation of the Dynamic Behavior of Brayton Batteries for Coupled Generation of Electricity, Heat, and Cooling

by Michael Krüger

Appl. Sci. 2025, 15(23), 12636; https://doi.org/10.3390/app152312636 (registering DOI) - 28 Nov 2025

Abstract

This study presents a comprehensive dynamic system analysis of air-based Brayton batteries for the coupled generation of electricity, heat, and cooling. Building upon a previously published structural concept study, the most promising system architectures were modeled and evaluated using quasi-stationary simulations with dynamically designed thermal energy storage (TES) in Ebsilon Professional^®. The results show round-trip efficiencies (RTEs) of up to 50% for pure electricity generation and round-trip utilizations (RTUs) exceeding 85% for combined heat and power. Integration of waste heat further increases RTU to more than 100%, albeit at the expense of electrical efficiency. Dynamic simulations demonstrate stable operation with load gradients up to 2 MW min⁻¹, highlighting suitability for flexible industrial and grid applications. Regenerator-based TES exhibits the most favorable trade-off between efficiency and cost, while hybrid configurations of solid and liquid media offer additional optimization potential. The estimated investment costs range between 200 and 800 EUR/kWh_el, comparable to other large-scale Carnot battery systems. The findings provide a validated framework for the techno-economic design and control of next-generation Brayton battery systems and lay the foundation for experimental validation and pilot-scale implementation. Full article

(This article belongs to the Section Energy Science and Technology)

►▼ Show Figures

Figure 1

16 pages, 7778 KB

Open AccessArticle

Possibilities of Using 3D Printing with Polymers as Structural Components

by Marcin Artur Kwapisz, Piotr Paszta and Wiktor Lacki

Materials 2025, 18(23), 5384; https://doi.org/10.3390/ma18235384 (registering DOI) - 28 Nov 2025

Abstract

This article investigates the potential of using polymer-based Fused Filament Fabrication (FFF) 3D printing to manufacture structural components. This study aimed to explore the application of this technology within manufacturing processes. This research focused on material analysis, using a gear wheel from a lawn mower’s drive system as a case study. To achieve the purpose of this study, a reverse engineering process was carried out to create 2D and 3D documentation based on a physical piece. The geometric model created in the CAD environment was optimized for volume. It was intended to keep the material expenditure unchanged, which is intensified during the 3D printing process due to the need to apply an adhesive layer and supports. The final design process of the geometric model of the prototype was subjected to numerical analyses in terms of total deformation and reduced stresses for materials commonly used as filaments in the FDM/FFF 3D printing technology. The basic filaments PLA, ABS, and PA6 and PA12 were analyzed in this study. The results of the analyses showed that two of the four selected filaments had to be rejected due to significant deterioration in strength properties. Finally, the prototype workpieces were printed using materials approved for the manufacturing process. Full article

(This article belongs to the Special Issue Advanced Smart Materials for 3D/4D Printing: Mechanisms, Performance and Applications)

►▼ Show Figures

Figure 1

15 pages, 1323 KB

Open AccessArticle

Quantitative Modeling of Fluid Phase Evolution in Ordovician Reservoirs of the Shuntuoguole Area, Tarim Basin: Implications for Oil and Gas Phase Differentiation

by Rui Deng, Chengsheng Chen and Yunpeng Wang

Energies 2025, 18(23), 6273; https://doi.org/10.3390/en18236273 (registering DOI) - 28 Nov 2025

Abstract

The Shuntuoguole area has become an important oil and gas exploration replacement zone; however, the complexity of its ultra-deep oil and gas phase behavior poses a challenge to petroleum exploration and development. Previous research lacks quantitative modeling of the evolution of the oil and gas phase. In this paper, the phase characteristics, phase evolution processes, and main factors influencing phase differentiation in Ordovician reservoirs of three typical wells in the Shuntuoguole area were studied quantitatively by integrating PVT simulation and basin modeling. The results indicate that the difference in geothermal field between the Shunbei, Shuntuo, and Shunnan areas profoundly influences oil and gas phase differentiation. The Ordovician fluid in well SB5 has remained in the oil phase since the light oil accumulated during the Late Hercynian period. In well MS1, the Ordovician fluid briefly existed in a gas–liquid two-phase state after the light oil accumulated in the Late Caledonian period, then transformed into the liquid phase at 348 Ma and has maintained this state to the present. In the dry gas reservoir of well SN5, the fluid has remained in a single gas phase throughout all stages of reservoir temperature and pressure evolution after accumulation. Full article

13 pages, 4729 KB

Open AccessArticle

Effect of Axial Stress on Torsional Behavior for Extruded AZ31 Mg Alloys Under Multiaxial Loading

by Chong Yang, Baocheng Yang, Guoguo Zhu, Muyu Li, Liangbin Chen, Yanpu Chao, Yaohui Li and Ruju Fang

Metals 2025, 15(12), 1316; https://doi.org/10.3390/met15121316 (registering DOI) - 28 Nov 2025

Abstract

The torsional behavior of an extruded AZ31 magnesium alloy was investigated by combined axial-torsion mechanical testing with different stress ratios. The SEM-EBSD was used to analyze the microstructure and texture evolution of deformed samples. The results indicate that the axial tension results in a concave-down shape of shear stress–strain curves, while a concave-up shape after yielding is presented during combined compression-torsion loading due to twinning mechanism. Compared to pure shear, the yield strength decreases by 7 MPa and shear strain increases by 1% under σ:τ = −1:1 with same shear stress. Due to the Swift effect, a strain partitioning is displayed for axial strain during combined tension-torsion loading with low stress ratio. The twin volume fraction is 90% under σ:τ = −1:1, and the local dislocation density with a KAM value of 1.1 is maximum under σ:τ = 2:1. The primary twin type is {10-12} twins with axial compression. The main deformation mode changes from basal slip to prismatic slip with increase in axial tension stress. Both basal slip and twinning are activated and the interaction between dislocation slip and twinning contributes to the complex strain hardening behavior during combined compression-torsion loading. Full article

30 pages, 4259 KB

Open AccessArticle

GRUAtt-Autoformer: A Hybrid Framework with BiGRU-Enhanced Attention for Crude Oil Price Forecasting

by Ying Zhang, Jie Wang and Ying Zhao

Mathematics 2025, 13(23), 3825; https://doi.org/10.3390/math13233825 (registering DOI) - 28 Nov 2025

Abstract

As a pivotal global commodity, crude oil price volatility directly impacts economic stability and strategic security. Being the most widely traded asset worldwide, it also serves as a key financial barometer and a critical transition fuel in the shift towards renewable energy. Nevertheless, accurate forecasting of crude oil prices remains challenging due to three persistent challenges: (1) the lack of a systematic method to filter out redundant and noisy features for deep learning models; (2) the limited ability of existing models to simultaneously capture both local bidirectional dependencies and global periodic patterns; and (3) the non-adaptive nature of conventional attention mechanisms, which restricts their capacity to dynamically focus on the most informative historical periods. To bridge these gaps, this study introduces a novel forecasting framework with three key contributions. First, we introduce a hierarchical feature selection paradigm based on LightGBM to systematically eliminate data redundancy and noise, thereby constructing an optimal feature subset for subsequent deep modeling. Second, an improved Autoformer encoder, integrated with Bidirectional GRUs, is designed to simultaneously capture local bidirectional dependencies and global periodic patterns, enabling a more comprehensive multi-scale temporal representation. Third, a dynamic fusion mechanism is incorporated to adaptively recalibrate the significance of historical timesteps. This enables the model to focus on periods rich in information, enhancing contextual awareness in predictions. Future research aims to enhance forecasting capabilities by achieving a deeper integration of local and global temporal representations, potentially through exploring advanced gating or sparse attention mechanisms. Full article

26 pages, 8112 KB

Open AccessArticle

Advancing Real-Time Aerial Wildfire Detection Through Plume Recognition and Knowledge Distillation

by Pirunthan Keerthinathan, Juan Sandino, Sutharsan Mahendren, Anuraj Uthayasooriyan, Julian Galvez, Grant Hamilton and Felipe Gonzalez

Drones 2025, 9(12), 827; https://doi.org/10.3390/drones9120827 (registering DOI) - 28 Nov 2025

Abstract

Uncrewed aerial systems (UAS)-based remote sensing and artificial intelligence (AI) analysis enable real-time wildfire or bushfire detection, facilitating early response to minimize damage and protect lives and property. However, their effectiveness is limited by three issues: distinguishing smoke from fog, the high cost of manual annotation, and the computational demands of large models. This study addresses the three key challenges by introducing plume as a new indicator to better distinguish smoke from similar visual elements, and by employing a hybrid annotation method using knowledge distillation (KD) to reduce expert labour and accelerate labelling. Additionally, it leverages lightweight YOLO Nano models trained with pseudo-labels generated from a fine-tuned teacher network to lower computational demands while maintaining high detection accuracy for real-time wildfire monitoring. Controlled pile burns in Canungra, QLD, Australia, were conducted to collect UAS-captured images over deciduous vegetation, which were subsequently augmented with the Flame2 dataset, which contains wildfire images of coniferous vegetation. A Grounding DINO model, fine-tuned using few-shot learning, served as the teacher network to generate pseudo-labels for a significant portion of the Flame2 dataset. These pseudo-labels were then used to train student networks consisting of YOLO Nano architectures, specifically versions 5, 8, and 11 (YOLOv5n, YOLOv8n, YOLOv11n). The experimental results show that YOLOv8n and YOLOv5n achieved an mAP@0.5 of 0.721. Plume detection outperforms smoke indicators (F1: 76.1–85.7% vs. 70%) in fog and wildfire scenarios. These findings underscore the value of incorporating plume as a distinct class and utilizing KD, both of which enhance detection accuracy and scalability, ultimately supporting more reliable and timelier wildfire monitoring and response. Full article

(This article belongs to the Special Issue Rapid Disaster Assessment and Post-Disaster Recovery Using UAV Remote Sensing)

Show export options Show export options

Select all

Export citation of selected articles as:

Error

Oops... you haven't selected anything for export.

Displaying article 1-50 on page 1 of 2000.

Go to page 1 2 3 4 5

Search Results (1,804,085)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI