Search Results (884)

Reliable cropland, forestland, and grassland maps support resource assessment and ecological management in arid and semi-arid Central Asia. Existing land-cover products often delineate these classes differently, vary in reliability across classes and locations, and may share the same errors even when they agree. This study formulates multi-product fusion as a pixel- and class-specific reliability decision problem. To address this problem, we propose a reliability-adaptive fusion framework, the Discrepancy-Aware Reliability-Adaptive Fusion Network (DRAFNet), using 2020 maps from three global 30 m land-cover products—FROM-GLC Plus, GLC-FCS30D, and GlobeLand30—and variables representing aridity, temperature, precipitation, elevation, and slope. Unlike fixed-weight fusion methods and segmentation models that use the source products only as input channels, DRAFNet retains the categorical source decisions and adjusts each contribution according to its estimated reliability for the assigned class and location. Weight removed from an unreliable source is transferred to a residual expert, which provides an alternative prediction when the source products are unreliable or share the same error. Voting entropy and geo-environmental variables provide contextual information for this decision. On independent test samples from the five Central Asian countries, DRAFNet achieved an overall accuracy (OA) of 0.8275, a Kappa coefficient of 0.7698, a mean intersection over union (mIoU) of 0.7046, and a macro-averaged F1 score (Macro F1) of 0.8241. These values were 0.95–1.38 percentage points higher than those of U-Net++, the strongest benchmark. Local comparisons indicated more coherent spatial patterns and clearer boundaries in areas of pronounced disagreement. The mean and median absolute log-ratio deviations from area statistics reported by the Food and Agriculture Organization of the United Nations (FAO) were 0.618 and 0.450, respectively, both lower than those of the source products. These results support land-resource assessment and ecological management in Central Asia. Full article

Unmanned aerial vehicle (UAV) imagery supports intelligent surveillance, environmental monitoring, traffic management, and infrastructure inspection. Yet aerial detection is difficult when objects are small, crowded, and observed at markedly different scales. Background clutter and illumination changes further weaken target cues and impair localization. We therefore propose HD-YOLO, a lightweight multiscale detector for small objects in UAV imagery. Its Multi-Dilation Shared Convolution Kernel (DSCK) extracts local texture and contextual information with shared dilated kernels. The Hybrid Dilated Bidirectional Feature Pyramid Network (HDFPN) reconstructs global and local cues before bidirectional aggregation, enabling high-resolution evidence to reach the prediction layers. The Efficient and Slim Head (ES-Head) combines shared operations with differential convolution to reduce cost and strengthen boundary-sensitive features. A joint ShapeIoU and Normalized Wasserstein Distance loss improves regression for small, irregular objects. Together, these components reduce missed detections in dense, cluttered scenes without relying on large model capacity. On VisDrone2019, HD-YOLO improves precision, recall, mAP50, and mAP50:95 over YOLOv8n by 6.9%, 7.2%, 8.2%, and 5.2%, respectively, while reducing parameters from 3.0 M to 0.9 M. Evaluations on TinyPerson and HIT-UAV also support its utility for tiny pedestrians and infrared aerial targets. HD-YOLO therefore improves small-object detection with a compact parameter footprint, while direct hardware benchmarks remain necessary to establish deployment efficiency. Full article

Early and accurate analysis of breast cancer is critical for improving patient outcomes. Ultrasound imaging is widely used for breast tumor screening due to its safety, accessibility, and low cost. We propose DGTN (Diffused Graph-Transformer Network), a lightweight, end-to-end deep learning model that jointly performs breast tumor segmentation and multi-class classification (benign, malignant, and normal) from ultrasound images. DGTN integrates Graph Convolutional Networks (GCNs) and Transformer encoders through a bidirectional diffusive attention mechanism and a learnable gating strategy, enabling structured spatial information and global contextual features to co-evolve. We evaluate DGTN on the public BUSI breast ultrasound dataset using balanced sampling and a joint cross-entropy and Dice loss. The model achieves 68.8% classification accuracy and a Dice score of 0.6227. While its performance remains below that of recent state-of-the-art pipelines, DGTN offers a favorable trade-off between accuracy and computational efficiency within a single unified framework. Ablation experiments indicate that both diffusive attention and gating contribute meaningfully to performance (paired t-test across five cross-validation folds, p < 0.05, with large paired effect sizes, d ≈ 1.0–1.4); because this test is based on only five folds, the result should be interpreted as indicative rather than conclusive, and we report it alongside fold-level effect sizes rather than as a stand-alone confirmation of significance. To the best of our knowledge, this work represents one of the first applications of diffusive graph-transformer co-learning to breast ultrasound imaging, demonstrating the potential of graph-enhanced attention for efficient multi-task medical image analysis. Full article

Timely and accurate termite detection is essential for effective termite control. To address the challenges posed by the small size of termite individuals and the susceptibility of target features to background texture interference under complex backgrounds, this study proposes ESD-YOLO, a fine-grained feature-enhanced object detection model. Using YOLO11n as the baseline, ESD-YOLO redesigns the feature extraction, deep feature aggregation, and multi-scale feature fusion stages to improve the representation of small-scale termite targets under complex backgrounds. Specifically, the Efficient Multi-scale Attention (EMA) mechanism is incorporated into the C3k2 module to enhance feature discriminability between termite individuals and the background. A Spatial Pyramid Pooling-Fast with Dual Global Pooling (SPPF-DGP) module is employed to supplement deep features with global contextual information and salient response information. In addition, the DySample dynamic upsampling module is introduced to improve spatial alignment during multi-scale feature fusion and enhance boundary representation for small targets. Experimental results show that ESD-YOLO achieves Precision, Recall, mAP@0.5, and mAP@0.5:0.95 values of 95.39%, 96.33%, 97.85%, and 65.92%, respectively, with 2.67 M parameters and 6.68 G FLOPs. Compared with Faster R-CNN, RetinaNet, RT-DETR, and several YOLO-series models, ESD-YOLO demonstrates strong small-target detection and localization performance under the controlled complex-background conditions established in this study, providing a methodological reference for automated termite detection in practical settings. Full article

Despite growing regulatory and investor attention to human capital as a core pillar of corporate sustainability and ESG reporting, it remains unclear how human capital-related information is priced in financial markets. This study examines the associations between human capital disclosure, management practices, and firms’ financing costs. Using a global panel of 1180 non-financial firms across 53 countries from 2017 to 2023, we employ Bloomberg ESG data to construct measures of human capital disclosure, management practices, and materiality. Panel regression analyses indicate that while human capital management practices exhibit no significant standalone associations, human capital disclosure is positively associated with the cost of equity, cost of debt, and the weighted average cost of capital. However, we document a significant complementary effect: when coupled with strong management practices, disclosure is associated with a lower cost of equity, particularly in contexts where human capital is financially material. Furthermore, the positive association between disclosure and capital costs becomes weaker as human capital materiality increases. Overall, our findings suggest that capital markets do not uniformly price human capital information; rather, its valuation is highly conditional, depending on the substantive credibility of management practices and contextual materiality. Full article

Background: Major depressive disorder (MDD) affects 332 million people worldwide, yet diagnosis remains reliant on subjective clinical interviews with substantial inter-rater variability. Objective neuroimaging model-attributed regions offer a path toward precision psychiatry, but existing computational approaches often lack clinical interpretability. Methods: We propose HDGNN-Mamba2, a Mamba-based spatiotemporal heterogeneous dynamic graph neural network. A hybrid Mamba2-GNN block with cross-attention fusion is developed to capture individual spatiotemporal contextual features and identify model-attributed regions. A heterogeneous global graph block with dynamic edge updating is constructed, integrating individual brain features with non-imaging phenotypic information (sex, age, education) to extract embeddings through inter-individual relationship modeling. Heterogeneous Graph Supervised Contrastive Learning is integrated to enhance discriminative capacity. Results: Evaluated on 533 subjects from the REST-meta-MDD dataset, HDGNN-Mamba2 achieved 83.88% accuracy, 86.52% sensitivity, and 80.85% specificity in ten-fold cross-validation. The identified model-attributed regions include the anterior cingulate cortex, parahippocampal gyrus, and thalamus. Conclusions: HDGNN-Mamba2 demonstrates competitive performance as an algorithmic framework for MDD classification, offering complementary architectural advantages in spatiotemporal fusion and interpretable region identification. Full article

Accurate teeth segmentation in panoramic dental radiographs remains a challenging task due to high image noise, low contrast, the similarity in intensity between teeth and surrounding tissues, and blurred tooth boundaries. To address these challenges, we propose DoubleTransU-Net, a dual-stage hybrid CNN–Transformer architecture that combines progressive segmentation refinement with global contextual feature learning. The first stage generates an initial tooth segmentation, while the second stage progressively refines ambiguous tooth regions to improve boundary delineation and segmentation accuracy. In addition, Atrous Spatial Pyramid Pooling (ASPP) modules capture multi-scale contextual information, whereas squeeze-and-excitation (SE) blocks enhance discriminative feature representations through channel-wise feature recalibration. The proposed model was evaluated on two public panoramic dental X-ray datasets, UFBA-UESC (1500 images) and Tufts (1000 images), and compared against several state-of-the-art segmentation models, including U-Net, DoubleU-Net, Attention U-Net, TransUNet, and DeepLabv3+. On the UFBA-UESC dataset, DoubleTransU-Net achieved an Accuracy of 95.54%, a Dice coefficient of 93.72%, an Intersection over Union (IoU) of 88.18%, a Precision of 93.57%, and a Recall of 94.24%. On the Tufts dataset, it achieved an Accuracy of 91.83%, a Dice coefficient of 92.93%, an IoU of 86.80%, a Precision of 92.27%, and a Recall of 93.97%. These results demonstrate that DoubleTransU-Net consistently outperforms existing state-of-the-art segmentation methods while exhibiting strong robustness and generalization across different panoramic dental datasets, highlighting its effectiveness for tooth semantic segmentation in panoramic dental X-ray images. Full article

Rice leaf diseases significantly affect crop health and yield potential, creating a need for accurate and timely disease diagnosis. Many existing approaches still rely on single-stream feature extraction architectures, which may limit the ability to simultaneously capture global contextual information and fine-grained disease characteristics. Moreover, limited interpretability and decision-support capability hinder their practical deployment in real-world rice farming. To address these limitations, we employed a framework consists of two parallel feature extraction streams designed to capture different characteristics of disease patterns. The first stream uses a Swin Transformer to learn global contextual information and long-range spatial relationships across the leaf image. The second stream employs ConvNeXt to extract local texture features, including lesion details, spots, and color variations. By combining these complementary representations, the proposed framework effectively integrates global semantic information with local disease-specific features for improved classification performance. The extracted features from the two streams are fused through concatenation followed by an attention-based feature refinement module, enabling adaptive weighting of discriminative features. The refined representation is then used by a fully connected classifier for disease prediction. To enhance model interpretability, Grad-CAM visualization is incorporated to highlight disease-relevant regions and provide visual explanations for the model decisions. Furthermore, an LLM-based advisory module is integrated as a post-diagnosis decision-support component to provide contextualized disease management information and suggestions based on the predicted disease category. The generated suggestions are intended to support, rather than replace, expert agronomic recommendations and should be validated by agricultural professionals before practical application. The proposed framework was evaluated on two rice leaf disease datasets, achieving accuracies of 99.55% and 97.06% on Dataset-1 and Dataset-2, respectively, which are higher than those reported in previous studies. Additionally, five-fold cross-validation on Dataset-1 achieved an average accuracy of 98.91% ± 0.47, demonstrating the stability of the proposed approach. Cross-dataset evaluation using nine common disease classes across both datasets achieved 90.80% accuracy, indicating improved generalization across different data distributions. The proposed framework provides an accurate and explainable approach for rice leaf disease diagnosis in smart agriculture applications. Full article

Early childhood dental caries (ECC) remains one of the most prevalent and preventable chronic diseases affecting young children globally, yet behavioral and contextual determinants of its onset during the first 1000 days of life remain poorly understood and infrequently modeled in Latin American populations. This study presents a novel application of the PRECEDE diagnostic framework to conceptualize Early Childhood Dental Caries during the first 1000 days (1000ECDC) as a behaviorally rooted, socio-ecologically conditioned health problem. Drawing on a formative longitudinal pilot study (n = 10 mother–infant dyads, Caracas, Venezuela) and a complementary narrative literature review, we develop the first PRECEDE-based conceptual model linking maternal prenatal and postnatal behaviors to dental caries risk in a Latin American context. The model identifies dietary and feeding behaviors, oral health care practices, and healthcare service utilization as primary behavioral risk factors, modulated by predisposing factors (low oral health knowledge, limited self-efficacy, cultural norms), enabling factors (socioeconomic constraints, fragmented health systems), and reinforcing factors (social norms, family influence). Exploratory pilot findings indicated that more than 60% of children had advanced dental caries lesions at 24-month follow-up, consistent with regional estimates and underscoring the urgency of early intervention. This model provides practitioners and policymakers with a structured, evidence-informed diagnostic tool to guide the design of early-life oral health promotion programs, with particular relevance for low-resource Latin American settings. Future validation through expert consensus and prospective studies is warranted. Full article

Small object detection (SOD) is a crucial research area in the field of computer vision. It poses significant challenges due to variations in scale, dense objects, limited target resolution, and a complex background. To achieve real-time detection, existing methods typically focus on local feature extraction and employ downsampling to reduce computation. However, this approach lowers the feature map resolution and loses the fine-grained details during downsampling. Meanwhile, frequency, global, and surrounding information play a significant role in small object detection. To address multi-scale dense targets in RGB and thermal infrared imagery with limited spatial, contextual, and frequency information, we propose a novel architecture that leverages Wavelet Transform Fusion (WTF) and Context-Guided Downsampling (CGD) in the real-time detection transformer (RT-DETR) for small object detection. WTF performs multi-frequency feature decomposition and fusion to preserve both high-frequency details and low-frequency semantic information, thereby improving the representation of small targets while reducing computational complexity. CGD incorporates local, surrounding, and global contextual information during downsampling to mitigate spatial information loss and strengthen feature representation for precise object localization. CGD is a downsampling technique that efficiently captures and preserves the contextual spatial information of local and global features using a local feature extractor and a joint feature extractor. It takes into account the surrounding information of the object, thereby reducing the loss of spatial details during downsampling. This spatial information helps in the precise detection of small objects in RGB and thermal infrared aerial images. Our proposed model is evaluated independently on the RGB aerial dataset NWPU-VHR-10 and the thermal infrared dataset HIT-UAV. Evaluations on the NWPU-VHR-10 and HIT-UAV datasets demonstrate that CGWT-DETR improves the mAP@0.50 to 89.9% and 86.5%, respectively, while boosting the strict localization metric mAP@0.50:0.95 to 60.3% and 58.8%. Furthermore, these accuracy gains are achieved alongside a 14.29% reduction in model parameters and a 29.8% decrease in GFLOPs. Experimental results demonstrate that CGWT-DETR outperforms the RT-DETR baseline in both detection accuracy and computational efficiency. Full article

Video action recognition requires the joint modeling of spatial appearance information and temporal dynamics. However, existing efficient action recognition methods based on two-dimensional convolution still have limitations in representing multi-scale spatial cues and aggregating key spatiotemporal information. To address these issues, this paper proposes an action recognition network based on Multi-Scale Dilated Feature Fusion and Decoupled Spatiotemporal Attention Pooling, termed MDSTA-Net. The proposed method adopts TSM as the basic temporal modeling framework and ResNet-50 as the backbone network. First, a Multi-Scale Dilated Feature Fusion module (MSDF) is designed to construct continuous multi-scale receptive fields through parallel convolutional branches with different dilation rates. An adaptive branch aggregation mechanism is further introduced to dynamically fuse responses at different scales, thereby enhancing the representation of both local details and broader contextual information. Second, a Decoupled Spatiotemporal Attention Pooling module (DSTAP) is proposed to model key action frames along the temporal dimension and salient discriminative regions along the spatial dimension. A residual pooling path is also incorporated to preserve global semantic information, improving the discriminative capability of video-level action representations. Experimental results on three public datasets, namely Something-Something V2, Kinetics-400, and HMDB51, demonstrate that MDSTA-Net achieves favorable recognition performance compared with several representative methods. Ablation studies further verify the effectiveness of MSDF and DSTAP, indicating that multi-scale spatial feature enhancement and key spatiotemporal information aggregation can effectively improve action recognition performance. Full article

Recently, remote sensing object detection (RSOD) still faces challenges, especially in scenarios involving multi-scale, small, and elongated objects. To address these issues, we propose the wavelet-enhanced cross-strip network, a novel backbone network specifically designed for RSOD. First, we design a Multi-scale Cross-Strip Convolution (MCSConv) module by integrating the local contextual capability of square convolutions with the anisotropic modeling advantages of strip convolutions, thereby forming a mixed square-strip receptive field. This design successfully captures directional features of elongated targets while significantly alleviating feature redundancy. Concurrently, its channel-adaptive weighting mechanism further enhances the representation capability for multi-scale targets. Furthermore, we introduce a High-Low Frequency Feature Enhancement (HLFE) module to decompose feature maps into low- and high-frequency components via the discrete Haar wavelet transform (DWT). A High-Frequency Denoising Enhancement (HFDE) module suppresses noise and enhances edge textures in high-frequency components to preserve crucial information of small objects in the shallow layers, while Gaussian convolution enhances low-frequency semantics to highlight global structures and object contours in the deep layers. Comprehensive evaluations on four benchmarks (DOTA-v1.0, DOTA-v1.5, DIOR-R, and FAIR1M1.0) verify that WECSNet achieves highly competitive detection performance with fewer parameters. Full article

Research on industrial location criteria has grown since the first industrial revolution, revealing key factors for choosing sites for industrial parks. However, macro-elements like economy, culture, and technology continuously reshape these criteria. Most existing studies focus on specific factors, indicating a need for a comprehensive synthesis. This study employs systematic literature review (SLR) methodology following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines. to analyze scientific information and aim to summarize essential criteria for selecting industrial zones. Following the screening and eligibility assessment process, 182 full-text studies were reviewed, with 60 recent publications emphasized for synthesizing current evidence, complemented by 56 supplementary references to support theoretical and contextual analysis. Findings include: (1) the need for frequent updates to criteria in response to global economic and scientific trends; (2) traditional factors like natural resources, transportation systems, costs, and population density remain significant; and (3) of the identified criteria organized into six main evaluation factors, public health emerged as an important one, especially highlighted by the COVID-19 pandemic. Given the potential for project failures due to overlooked criteria, regularly reviewing and updating industrial location considerations is essential and can serve as a valuable resource for future research and practice. Full article

Computer-aided diagnostic technologies have demonstrated substantial advantages in 3D medical image segmentation, particularly in multimodal 3D medical image segmentation tasks, where they play a pivotal role in driving continuous innovation in related architectures. As an integration of U-Net and Transformer, the UNETR architecture has demonstrated remarkable efficacy in 3D medical image segmentation. Nevertheless, despite its successes, UNETR remains challenged by clinical complexities such as intricate tumor localization and anatomical structural diversity in complex clinical settings. To address these issues, we propose an enhanced 3D segmentation framework, UAtten-Unetr, designed to improve segmentation accuracy and robustness in complex medical scenarios. The framework captures global contextual information via hierarchical Transformer layers and incorporates a spatial–channel attention module to enable adaptive fusion of multimodal features, thereby effectively enhancing cross-modal feature alignment capabilities. Concurrently, we innovatively developed a unified loss function based on bimodal modality-specific Dice constraints and uncertainty regularization, optimized for synchronous learning across the ACDC (cardiac MRI) and AMOS22 (abdominal CT/MRI) datasets. Experimental results showed that UAtten-Unetr achieved an average Dice score of 92.20% on the ACDC dataset, exceeding the reported nnU-Net result of 91.61% by 0.59 percentage points. On the AMOS22 dataset, the proposed method achieved an average Dice score of 84.51%, exceeding the reported UNETR result of 78.33% by 6.18 percentage points. However, its myocardium Dice score (84.11%) was lower than those of nnU-Net (89.24%) and MT-UNet (89.04%), indicating a remaining limitation in myocardium boundary segmentation. These results indicate competitive segmentation performance under the reported experimental settings. This method delivers dual improvements in accuracy and generalization across complex anatomical scenarios, providing an effective solution for precise diagnosis in intricate clinical environments. Full article

This study proposes a conceptual model of smart innovation to examine internationalization processes in traditional, territorially embedded industries, with a focal empirical focus on Sogrape, Portugal’s leading wine company, situated within the Vinho Verde wine sector. While digital transformation is increasingly recognized as a driver of competitiveness, its role in shaping internationalization strategies within regional agri-food systems remains insufficiently theorized. To address this gap, the study adopts a modified Delphi-based qualitative approach involving a panel of experts composed mainly of Sogrape managers and complemented by independent producers from the Vinho Verde wine sector. Through two iterative rounds of structured expert inquiry, the research identifies key mechanisms linking digital transformation, organizational capabilities, territorial identity, and international market expansion. The findings are synthesized into an integrative conceptual model that articulates how smart innovation, understood as the strategic alignment of digital capabilities, organizational processes, and territorial assets, may support internationalization processes in territorially embedded settings. The model emphasizes the role of digital platforms, data-driven decision-making, and narrative-driven place positioning in translating territorial identity into competitive value in global markets. Importantly, the study does not claim to provide representative evidence of the Vinho Verde wine sector as a whole. Rather, it develops a focal-case-based analytical architecture, grounded primarily in Sogrape’s organizational context and qualified by complementary insights from independent producers. The study contributes to the literature by bridging digital transformation, internationalization, and territorial value creation within a unified conceptual framework. From a managerial perspective, it offers analytically grounded insights that may inform strategic reflection in wine firms and other territorially embedded agri-food sectors, subject to contextual adaptation and further empirical validation. Full article