Saved Queries

This study explores the adoption of AI in the UAE’s creative industries through interviews with nine professionals, primarily Emiratis, from journalism, filmmaking, content creation, and heritage sectors. Their insights shed light on the intricate relationship between technological progress, job stability, and the preservation of cultural integrity. One of the key observations was the dual nature of AI’s impact: while it undeniably enhances operational efficiency and reduces expenses, there are valid concerns regarding the authenticity and quality of AI-generated content and its potential impact on the development and utilization of professional skills within these industries. Despite the UAE government’s strategic initiatives to promote AI adoption, the findings revealed a notable absence of clear guidelines, placing the onus on individuals to proactively navigate the landscape of AI integration. This research challenges the prevailing narratives that often depict the Global South as passive consumers of technology as it highlights the participants’ acute awareness of the inherent biases present in AI technologies, particularly in the representation of their local culture. Full article

(This article belongs to the Special Issue Unravelling the Media’s Role in Technological Innovation and AI's Environmental, Social, and Economic Impacts)

16 pages, 5435 KiB

Open AccessArticle

PAPRec: 3D Point Cloud Reconstruction Based on Prior-Guided Adaptive Probabilistic Network

by Caixia Liu, Minhong Zhu, Yali Chen, Xiulan Wei and Haisheng Li

Sensors 2025, 25(5), 1354; https://doi.org/10.3390/s25051354 (registering DOI) - 22 Feb 2025

Abstract

Inferring a complete 3D shape from a single-view image is an ill-posed problem. The proposed methods often have problems such as insufficient feature expression, unstable training and limited constraints, resulting in a low accuracy and ambiguity reconstruction. To address these problems, we propose a prior-guided adaptive probabilistic network for single-view 3D reconstruction, called PAPRec. In the training stage, PAPRec encodes a single-view image and its corresponding 3D prior into image feature distribution and point cloud feature distribution, respectively. PAPRec then utilizes a latent normalizing flow to fit the two distributions and obtains a latent vector with rich cues. PAPRec finally introduces an adaptive probabilistic network consisting of a shape normalizing flow and a diffusion model in order to decode the latent vector as a complete 3D point cloud. Unlike the proposed methods, PAPRec fully learns the global and local features of objects by innovatively integrating 3D prior guidance and the adaptive probability network under the optimization of a loss function combining prior, flow and diffusion losses. The experimental results on the public ShapeNet dataset show that PAPRec, on average, improves CD by 2.62%, EMD by 5.99% and F1 by 4.41%, in comparison to several state-of-the-art methods. Full article

(This article belongs to the Special Issue Artificial Intelligence in Computer Vision: Methods and Applications—2nd Edition)

►▼ Show Figures

Figure 1

22 pages, 1543 KiB

Open AccessReview

A Brief Review of Multi-Physics Coupling Research on Hydroelectric Generators

by Jiwen Zhang, Xingxing Huang and Zhengwei Wang

Energies 2025, 18(5), 1074; https://doi.org/10.3390/en18051074 (registering DOI) - 22 Feb 2025

Abstract

Hydropower, with its high degree of flexibility, plays an important role in the transformation of the global energy mix. Generators are the core component of the hydropower units; their performance directly affects the efficiency and reliability of the hydroelectric units. The dynamic characteristics of a generator during operation are usually the result of the coupling and interaction of multiple physical fields. Therefore, the interactions among electromagnetic, thermal, structural, and fluid fields inside hydroelectric generators have become of great concern. This paper briefly reviews the hydroelectric generator multi-physics coupling investigations, which include research conducted through field measurements, theoretical analysis, and numerical simulations. The review covers electromagnetic vibrations of generators under the influence of electromagnetic and structural fields, heat generation of generators under the influence of electromagnetic and thermal fields, ventilation and heat dissipation of generators under the influence of flow and thermal fields, and physical field changes of generators under the influence of electromechanical signals. The review also highlights unresolved issues in the field of hydropower that could benefit from fundamental research using a multi-physics coupling approach. Full article

(This article belongs to the Section A: Sustainable Energy)

►▼ Show Figures

Figure 1

20 pages, 2021 KiB

Open AccessArticle

A Novel Metabolic Risk Classification System Incorporating Body Fat, Waist Circumference, and Muscle Strength

by Carlos Raúl Robledo-Millán, María Regina Diaz-Domínguez, Ari Evelyn Castañeda-Ramírez, Efren Quiñones-Lara, Sebastián Valencia-Marín, Ricardo Xopán Suárez-García, Nely Gisela López-Desiderio, Claudio Adrián Ramos-Cortés, Areli Marlene Gaytán-Gómez, Juan Manuel Bello-López and Héctor Iván Saldívar-Cerón

J. Funct. Morphol. Kinesiol. 2025, 10(1), 72; https://doi.org/10.3390/jfmk10010072 (registering DOI) - 22 Feb 2025

Abstract

Background: As metabolic diseases continue to rise globally, there is a growing need to improve risk assessment strategies beyond traditional measures such as BMI and waist circumference, which may fail to identify individuals at risk. This study develops and validates a novel metabolic risk classification system that incorporates body fat percentage (%BF), waist circumference (WC), and grip strength (GS) in Mexican adults. It aims to improve risk stratification and evaluate the association with metabolic syndrome. Methods: This cross-sectional study involved 300 young adults (18–22 years) from a university in Mexico City, utilizing body composition (%BF) and anthropometric measures (WC, GS) to categorize them into four risk groups: protective, low risk, increased risk, and high risk. A retrospective cohort of 166 adults (18–65 years) with complete clinical records was used for validation. Results: The inclusion of GS in the risk assessment significantly shifted the distribution in the young adult cohort, reducing the “no risk” category (15.5% males, 11.6% females) and expanding the higher-risk categories (70.2% males, 69% females). Metabolic parameters such as fasting glucose, triglycerides, HDL cholesterol, and blood pressure worsened progressively across the risk categories (p < 0.001). The high-risk group exhibited a markedly increased odds ratio for metabolic syndrome at 28.23 (10.83–73.6, p < 0.001), with no cases in the protective and low-risk groups. Conclusions: Integrating grip strength with %BF and WC into a risk classification system substantially enhances metabolic risk stratification, identifies at-risk individuals not previously detected, and confirms a protective group. This validated system provides a robust tool for early detection and targeted interventions, improving public health outcomes in metabolic health. Full article

(This article belongs to the Special Issue Physical Activity for Optimal Health)

44 pages, 14529 KiB

Open AccessArticle

Investigation of Carbon Monoxide, Carbon Dioxide, and Methane Source Variability at the WMO/GAW Station of Lamezia Terme (Calabria, Southern Italy) Using the Ratio of Ozone to Nitrogen Oxides as a Proximity Indicator

by Francesco D’Amico, Teresa Lo Feudo, Daniel Gullì, Ivano Ammoscato, Mariafrancesca De Pino, Luana Malacaria, Salvatore Sinopoli, Giorgia De Benedetto and Claudia Roberta Calidonna

Atmosphere 2025, 16(3), 251; https://doi.org/10.3390/atmos16030251 (registering DOI) - 22 Feb 2025

Abstract

In the field of Atmospheric Sciences, source apportionment and a more detailed understanding of local and remote contributions to observed concentrations of greenhouse gases (GHGs) across international networks, such as the World Meteorological Organization—Global Atmosphere Watch (WMO/GAW), can be achieved via the implementation of new atmospheric tracers. One tool for achieving a more precise understanding of GHG emissions is the evaluation of air mass aging indicators, which can serve as proximity indicators. In this study, the ratio between ozone (O₃) and nitrogen oxides (NO_x) is applied to nine continuous years (2015–2023) of measurements at the Lamezia Terme (LMT) observation site in Calabria, Southern Italy, to differentiate the aging of air masses and identify four distinct categories: LOC (local), N–SRC (near source), R–SRC (remote source), and BKG (atmospheric background). Due to possible overestimation of nitrogen dioxide (NO₂) caused by heated (~300–400 °C) molybdenum converters used in the employed instruments, a correction factor based on a previous study has been integrated to further analyze the results. Additionally, this work introduces a second correction factor based on the local behavior of surface ozone and the diurnal peaks observed during boreal warm seasons in an area characterized by a Mediterranean climate. The results of this study confirm the hypotheses of previous works on local sources of pollution: the LOC category yields the highest concentrations observed at the site, which are in accordance with the northeastern wind sector and anthropogenic sources. R–SRC and BKG are more representative of atmospheric background levels and characterize westerly winds from the Tyrrhenian Sea. N–SRC, as expected, shows an intermediate behavior between local and remote/background levels. Differences in results between standard O₃/NO_x categories and corrected measurements will need to be investigated in future studies. Full article

(This article belongs to the Section Air Pollution Control)

28 pages, 2187 KiB

Open AccessArticle

Online Review-Assisted Product Improvement Attribute Extraction and Prioritization Method for Small- and Medium-Sized Enterprises

by Keqin Wang, Angqi Lei, Zhihong Huang, Zhijiao Gao, Qingyu Ma, Chen Zheng, Jing Li, Benoît Eynard and Jinhua Xiao

Systems 2025, 13(3), 149; https://doi.org/10.3390/systems13030149 (registering DOI) - 22 Feb 2025

Abstract

Small- and medium-sized enterprises (SMEs) play a vital role in the global economy, driving innovation and economic growth, despite constraints on their financial and operational resources. In the competitive landscape of modern markets, continuous product design improvement has become essential for SMEs to meet dynamic user requirements, enhance satisfaction, and maintain competitiveness. Online reviews have emerged as valuable sources of user feedback, offering real-time, large-scale insights into user preferences. However, existing methods for leveraging online reviews in product design improvement have significant limitations, including insufficient attention paid to the hierarchical structure of different attributes when extracting product improvement attributes and a lack of quantitative attribute prioritization strategies. These shortcomings often result in suboptimal improvement and inefficient resource allocation, particularly for SMEs with limited resources. To address these challenges, this study proposed a novel online review-assisted method for product design improvement tailored to the needs of SMEs. The proposed method incorporates a hierarchical latent Dirichlet allocation model to extract and organize product attributes hierarchically, thereby enabling a comprehensive understanding of user requirements. Furthermore, a marginal utility-based approach is employed to prioritize product improvement attributes quantitatively, ensuring that the most impactful attributes are addressed efficiently. The effectiveness of the proposed method was demonstrated through a case study on the design improvement of a robotic vacuum cleaner developed using a typical SME in robotic cleaning solutions. Full article

(This article belongs to the Section Systems Engineering)

15 pages, 1567 KiB

Open AccessArticle

Phenotypic and Molecular Characterization of Pyomelanin-Producing Acinetobacter baumannii ST2_Pas;ST1816/ST195_Oxf Causing the First European Nosocomial Outbreak

by Alessandro Leonildi, Alfredo Rosellini, Giulia Gemignani, Giusy Tiseo, Marco Falcone, Cesira Giordano and Simona Barnini

Microorganisms 2025, 13(3), 493; https://doi.org/10.3390/microorganisms13030493 (registering DOI) - 22 Feb 2025

Abstract

Acinetobacter baumannii is one of the most successful and feared nosocomial pathogens. A. baumannii is considered a global threat in the healthcare setting, mainly owing to its ability to acquire multidrug resistance phenotypes. The A. baumannii pathogenesis is guided by its environmental persistence, as well as the production of numerous virulence factors. In several bacteria, the production of pigments, such as melanin, has indeed been linked with virulence and pathogenicity. Melanin is a brownish pigment, rarely observed in A. baumannii, that potentially reduces the susceptibility of the bacteria to host defense mechanisms and environmental insults. This study reports the first outbreak in Europe by pyomelanin-producing A. baumannii strains, in a tertiary-care university hospital in Pisa, Italy. Phenotypic and molecular analyses were performed. Full article

(This article belongs to the Collection Feature Papers in Medical Microbiology)

►▼ Show Figures

Figure 1

19 pages, 1050 KiB

Open AccessReview

Circulating Tumour DNA for Ovarian Cancer Diagnosis and Treatment Monitoring: What Perspectives for Clinical Use?

by Du-Bois Asante, Domenico Tierno, Gabriele Grassi and Bruna Scaggiante

Int. J. Mol. Sci. 2025, 26(5), 1889; https://doi.org/10.3390/ijms26051889 (registering DOI) - 22 Feb 2025

Abstract

Globally, ovarian cancer (OC) is the eighth most common malignant tumour in women. Unfortunately, its symptoms—especially at the early stages—are vague and non-specific, and, thus, most patients are diagnosed at the advanced stages of the disease (stage III and IV) when treatment is not curative. The currently available approved biomarkers are not sufficient for effective screening, prognosis, or monitoring of OC. Liquid biopsy tests such as circulating tumour DNA (ctDNA) analysis has the advantage of monitoring response to treatment in real time and providing a comprehensive genotypic profile of primary, metastatic, and recurrent tumours. Thus, ctDNA analysis can be used as a complementary test for effective diagnosis and monitoring of OC. We comprehensively review current studies (2019–2024) on OC, critically highlighting recent developments and applications of ctDNA for the diagnosis and management of the disease. Full article

(This article belongs to the Section Molecular Oncology)

►▼ Show Figures

Figure 1

17 pages, 1723 KiB

Open AccessSystematic Review

Quality of Life After Locoregional Treatment in Women with De Novo Metastatic Breast Cancer: A Systematic Review and Meta-Analysis

by Camille Weiss, Philippe Trensz, Martin Schmitt and Massimo Lodi

Cancers 2025, 17(5), 751; https://doi.org/10.3390/cancers17050751 (registering DOI) - 22 Feb 2025

Abstract

Introduction: Primary site locoregional treatment (LRT) of metastatic breast cancer has been performed and evaluated with the aim to improve survival, prevent complications, and alleviate local symptoms. As some studies fail to show a survival benefit, the quality of life is important to consider when deciding on LRT. The aim of this study was to evaluate and quantify the impact of LRT on the quality of life of patients with de novo metastatic breast cancer (dnMBC) through a systematic review of the literature and a meta-analysis. Methods: Multiple databases were searched on May 2024 with the following keywords: (i) dnMBC; (ii) LRT, including surgery +/− radiotherapy; and (iii) QOL. Results: Six studies were included in the qualitative synthesis and four in meta-analysis (481 women, n = 251 in the LRT and n = 230 in the control groups). There was a significant QOL decrease in the LRT group at 18 months (standardized mean difference [SMD] = −0.63; 95% confidence interval [CI] −0.98–−0.26; p < 0.001, low heterogeneity I² = 33%) and after 30 months (SMD −0.82; 95%CI −1.58–−0.06; p = 0.034, high heterogeneity I² = 93%), while no statistically significant difference was observed at short term (6 months, p = 0.333). Conclusions: This study shows that there is lacking evidence regarding the QOL benefits after LRT in this population, and even a numerical deterioration in global QOL several months after the treatment. Future and ongoing research may provide additional insights into this question on dnMBC and specifics subgroups. Full article

(This article belongs to the Special Issue New Developments in Breast Cancer Surgery, Risk, and Related Quality of Life)

►▼ Show Figures

Figure 1

21 pages, 577 KiB

Open AccessReview

Bioremediation of Endocrine Disruptors (EDs): A Systematic Review of Fungal Application in ED Removal from Wastewater

by Camila Emanuelle Mendonça Viana, Valquíria dos Santos Lima, Kelly Rodrigues, Luciana Pereira and Glória Maria Marinho Silva

Water 2025, 17(5), 640; https://doi.org/10.3390/w17050640 (registering DOI) - 22 Feb 2025

Abstract

Endocrine disruptors (EDs), including natural estrogens, such as 17β-estradiol (E2) and synthetic chemicals (e.g., bisphenol A (BPA) and per- and polyfluoroalkyl substances (PFAS)), pose environmental and human health risks due to their ability to interfere with hormonal systems, even at trace concentrations and can lead to developmental, reproductive, and carcinogenic effects. These persistent compounds often escape removal in conventional wastewater treatment processes, leading to environmental contamination and human exposure. Given their widespread presence in wastewater and resistance to conventional treatments, the use of fungi offers a promising bioremediation strategy. This review explores the potential of fungal biodegradation, particularly using the white-rot fungus Trametes versicolor, in mitigating the estrogenic activity of EDs in wastewater. Laccase, an oxidative enzyme produced by white-rot fungus, shows high efficiency in degrading EDs, positioning fungal treatment as an eco-friendly alternative to conventional technologies. This systematic literature review was conducted using the Methodi Ordinatio, a multi-criteria decision-making methodology that allows for a structured selection of relevant studies and underscores the significant potential of fungal-based systems in addressing the global challenge of ED contamination in water environments. Full article

(This article belongs to the Special Issue Biological Treatment of Water Contaminants: A New Insight)

13 pages, 807 KiB

Open AccessArticle

Association Between Joint Pain and Cancer in 8.45 Million Korean Adults: Insights from a National Cross-Sectional Study

by Taewook Kim

J. Clin. Med. 2025, 14(5), 1478; https://doi.org/10.3390/jcm14051478 (registering DOI) - 22 Feb 2025

Abstract

Background: Joint pain, a multifactorial musculoskeletal symptom, is rising globally due to an aging population. Simultaneously, cancer is increasingly considered a chronic condition with growing prevalence and improved survival rates, similar to hypertension and diabetes. Although the association between chronic diseases such as diabetes and joint pain has been well studied, the relationship between cancer and joint pain remains underexplored, especially as cancer’s chronic disease status evolves. Methods: This study analyzed data from the Korean National Health and Nutrition Examination Survey (KNHANES V) to investigate associations between cancer and joint pain in 8,451,047 individuals, representing Koreans over 50. Descriptive analyses identified demographic characteristics and disparities in joint pain prevalence by age and sex. Multivariate logistic regression analyzed seven common cancers in relation to spine, hip, and knee pain, adjusting for various factors and the Kellgren–Lawrence radiographic grade to pinpoint cancers significantly associated with each joint pain type. Results: Analysis demonstrated significant associations between certain cancers and joint pain. Back pain was linked to gastric, liver, cervical, and lung cancers; hip pain to breast and thyroid cancers; and knee pain to liver cancer. These findings underline complex relationships that suggest further investigation is needed to clarify specific cancer-related joint pain mechanisms. Conclusions: Descriptive and regression analyses highlighted essential demographic factors and significant associations between certain cancers and joint pain types. These insights enhance understanding of cancer’s chronic impact on joint pain and underscore the need for further research to refine these associations. Full article

(This article belongs to the Section Orthopedics)

►▼ Show Figures

Figure 1

13 pages, 1175 KiB

Open AccessArticle

Content-Based Histopathological Image Retrieval

by Camilo Nuñez-Fernández , Humberto Farias and Mauricio Solar

Sensors 2025, 25(5), 1350; https://doi.org/10.3390/s25051350 (registering DOI) - 22 Feb 2025

Abstract

Feature descriptors in histopathological images are an important challenge for the implementation of Content-Based Image Retrieval (CBIR) systems, an essential tool to support pathologists. Deep learning models like Convolutional Neural Networks and Vision Transformers improve the extraction of these feature descriptors. These models typically generate embeddings by leveraging deeper single-scale linear layers or advanced pooling layers. However, these embeddings, by focusing on local spatial details at a single scale, miss out on the richer spatial context from earlier layers. This gap suggests the development of methods that incorporate multi-scale information to enhance the depth and utility of feature descriptors in histopathological image analysis. In this work, we propose the Local–Global Feature Fusion Embedding Model. This proposal is composed of three elements: (1) a pre-trained backbone for feature extraction from multi-scales, (2) a neck branch for local–global feature fusion, and (3) a Generalized Mean (GeM)-based pooling head for feature descriptors. Based on our experiments, the model’s neck and head were trained on ImageNet-1k and PanNuke datasets employing the Sub-center ArcFace loss and compared with the state-of-the-art Kimia Path24C dataset for histopathological image retrieval, achieving a Recall@1 of 99.40% for test patches. Full article

(This article belongs to the Special Issue Deep Learning-Based Image and Signal Sensing and Processing: 2nd Edition)

10 pages, 242 KiB

Open AccessReview

Genetics of Gallstones

by Agnieszka Pęczuła, Adam Czaplicki and Adam Przybyłkowski

Genes 2025, 16(3), 256; https://doi.org/10.3390/genes16030256 (registering DOI) - 22 Feb 2025

Abstract

Gallstone disease (GSD) is a common gastrointestinal disorder affecting approximately 10–20% of the global adult population, characterized by the presence of gallstones, predominantly cholesterol-based, in the gallbladder and/or biliary ducts. While many patients remain asymptomatic, more than 20% develop clinical symptoms such as abdominal pain, nausea, vomiting, jaundice, and anorexia, potentially leading to severe complications like acute cholecystitis and biliary pancreatitis. GSD has a significant genetic predisposition, with the variable prevalence of the disease according to ethnicity being highest in American and European countries and lowest in Asian and African populations. Numerous genes encoding membrane transporters involved in bile metabolism are associated with GSD, including in particular members of ATP-binding cassette transporters and others, which affect bile lithogenicity and contribute to the development of gallstones. Specific mutations in these genes are linked to an increased risk of gallstone formation, especially in individuals with certain hereditary conditions such as hemolytic diseases, thyroid disorders, and hyperparathyroidism. Advances in genetic studies have identified new variants that influence the risk of cholelithiasis, although the exact mechanisms remain partially understood in many cases. This review briefly summarizes the genetic causes of cholelithiasis, highlighting various pathogenetic mechanisms. It presents the currently used treatments and the potential implications of widely applied genetic diagnostics. Full article

(This article belongs to the Special Issue Feature Papers in Human Genomics and Genetic Diseases 2024)

14 pages, 7475 KiB

Open AccessArticle

Therapeutic Effects of DNase I on Peripheral and Local Markers of Liver Injury and Neutrophil Extracellular Traps in a Model of Alcohol-Related Liver Disease

by Paulína Belvončíková, Andrej Feješ, Barbora Gromová, Ľubica Janovičová, Anna Farkašová, Pavel Babál and Roman Gardlík

Int. J. Mol. Sci. 2025, 26(5), 1893; https://doi.org/10.3390/ijms26051893 (registering DOI) - 22 Feb 2025

Abstract

Alcohol-related liver disease (ALD) is a leading cause of chronic liver conditions globally. Chronic alcohol consumption induces liver damage through various mechanisms, including neutrophil extracellular trap (NET) formation. Extracellular DNA (ecDNA), released from damaged hepatocytes and NETotic neutrophils, has emerged as a potential biomarker and contributor to liver disease pathology. Enzyme DNases could be an effective therapy for the denaturation of immunogenic ecDNA. This study investigated the circulating ecDNA and NET markers in ALD and therapeutic effect of DNase I in a murine model of ALD. Female C57BL/6J mice were fed a control diet (n = 13) or Lieber–DeCarli ethanol diet for 10 days followed by a binge ethanol dose to mimic acute-on-chronic alcoholic liver injury. From day 5, mice fed ethanol were randomized into an ethanol diet group (n = 17) and ethanol + DNase group (n = 5), which received additional DNase I treatment every 12 h. Liver damage markers were analyzed. Circulating ecDNA and NETosis were measured by fluorometry and cytometry, respectively. DNase I activity was analyzed with single radial enzyme dispersion assay. The ethanol-fed mice exhibited increased mortality, neutrophil infiltration and structural damage in the liver. Total circulating ecDNA levels and NET markers did not differ between groups. DNase activity was higher in ethanol-fed mice compared to controls and additional daily administration of DNase prevented liver injury. These findings suggest that alcohol-induced liver injury modestly influences systemic NETosis and ecDNA levels. However, increased DNase activity can prevent disease progression and enhanced systemic degradation of ecDNA using DNase I. Full article

(This article belongs to the Special Issue Molecular Advances and Insights into Liver Diseases)

►▼ Show Figures

Figure 1

23 pages, 29156 KiB

Open AccessArticle

U-MGA: A Multi-Module Unet Optimized with Multi-Scale Global Attention Mechanisms for Fine-Grained Segmentation of Cultivated Areas

by Yun Chen, Yiheng Xie, Weiyuan Yao, Yu Zhang, Xinhong Wang, Yanli Yang and Lingli Tang

Remote Sens. 2025, 17(5), 760; https://doi.org/10.3390/rs17050760 (registering DOI) - 22 Feb 2025

Abstract

Arable land is fundamental to agricultural production and a crucial component of ecosystems. However, its complex texture and distribution in remote sensing images make it susceptible to interference from other land cover types, such as water bodies, roads, and buildings, complicating accurate identification. Building on previous research, this study proposes an efficient and lightweight CNN-based network, U-MGA, to address the challenges of feature similarity between arable and non-arable areas, insufficient fine-grained feature extraction, and the underutilization of multi-scale information. Specifically, a Multi-Scale Adaptive Segmentation (MSAS) is designed during the feature extraction phase to provide multi-scale and multi-feature information, supporting the model’s feature reconstruction stage. In the reconstruction phase, the introduction of the Multi-Scale Contextual Module (MCM) and Group Aggregation Bridge (GAB) significantly enhances the efficiency and accuracy of multi-scale and fine-grained feature utilization. The experiments conducted on an arable land dataset based on GF-2 imagery and a publicly available dataset show that U-MGA outperforms mainstream networks (Unet, A2FPN, Segformer, FTUnetformer, DCSwin, and TransUnet) across six evaluation metrics (Overall Accuracy (OA), Precision, Recall, F1-score, Intersection-over-Union (IoU), and Kappa coefficient). Thus, this study provides an efficient and precise solution for the arable land recognition task, which is of significant importance for agricultural resource monitoring and ecological environmental protection. Full article

►▼ Show Figures

Figure 1

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 (112,387)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI