Search Results (2,020)

In the context of global climate governance and China’s ‘Dual Carbon’ strategy, enhancing carbon emission efficiency (CEE) is a critical pathway toward high-quality development. Urban–rural integration (URI), reshaping urban–rural structures and resource allocation, has significant environmental implications. However, the mechanisms through which URI influences city-level CEE remain underexplored. Using panel data from 283 Chinese prefecture-level cities (2005–2022), we employ a Spatial Durbin Model to investigate URI’s direct and spatial spillover effects. First, spatiotemporally, URI demonstrates an imbalanced pattern, with higher levels in eastern coastal regions and lower levels in central and western areas. Conversely, CEE exhibits a north–south divide, with higher efficiency in the south. URI advancement has been sluggish with persisting imbalances, whereas CEE has demonstrated a consistent upward trend. Second, the relationship between URI and CEE is characterized by nonlinearity and spatial dependence. The direct effect follows a U-shaped curve, initially inhibiting but later promoting local CEE once a threshold is surpassed (URI = 0.103). The spatial spillover effect follows an inverted U-shaped trajectory (threshold URI = 0.179), suggesting that inter-regional dynamics evolve from synergistic promotion to potential competition. These findings underscore the necessity of phased, adaptive policies to unlock the potential between URI and CEE, providing a scientific basis for coordinating urban–rural development with carbon neutrality objectives. Full article

Monocular depth estimation is a fundamental task with broad applications in autonomous driving and augmented reality. While recent lightweight methods achieve impressive performance, they often neglect the interaction of mid-order semantic features, which are crucial for capturing object structures and spatial relationships that directly impact depth accuracy. To address this limitation, we propose MogaDepth, a lightweight yet expressive architecture. It introduces a novel Continuous Multi-Order Gated Aggregation (CMOGA) module that explicitly enhances mid-level feature representations through multi-order receptive fields. In addition, we present MambaSync, a global–local interaction unit that enables efficient feature communication across different contexts. Extensive experiments demonstrate that MogaDepth achieves highly competitive or superior performance on KITTI, improving key error metrics while maintaining comparable model size. On the Make3D benchmark, it consistently outperforms existing methods, showing strong robustness to domain shifts and challenging scenarios such as low-texture regions. Moreover, MogaDepth achieves an improved trade-off between accuracy and efficiency, running up to 13% faster on edge devices without compromising performance. These results establish MogaDepth as an effective and efficient solution for real-world monocular depth estimation. Full article

Global Navigation Satellite System Reflectometry (GNSS-R) provides a vital means for sea surface wind speed retrieval, yet its application under extreme typhoon conditions remains challenging. Conventional geophysical models (GMFs) saturate in high wind speed regimes (>20 m/s), and deep learning models (e.g., CNNs) are constrained by data sparsity and feature complexity in typhoon environments. To address these issues, we propose a Comparative Learning method of CNN-Transformer with GMF fusion (CLCTG). The CNN branch extracts local coupling patterns, the Transformer branch models global dependencies, and Kullback–Leibler (KL) divergence loss is used for contrastive learning to heighten sensitivity to complex typhoon wind fields. The GMF branch serves as a physical reference/anchor in the low- to moderate-wind-speed range (<20 m/s) to guide the learning of data-driven branches and avoid overfitting by any single data-driven path. The adaptive fusion branch dynamically reweights the three branch outputs, combining local statistical characteristics to improve performance over approximately 0–30 m/s and extending the range of reliable GNSS-R retrieval from about 20 m/s to about 30 m/s; it should be noted that CLCTG exhibits a performance bottleneck in the extreme >30 m/s range. To further improve high-wind-speed predictions, we introduce environmental features based on their correlation with wind speed; ablation experiments demonstrate that the combined use of environmental parameters and CYGNSS features maximizes overall accuracy. Testing on five typhoons from the Eastern and Western Hemispheres confirms CLCTG’s generalization across diverse geographic contexts, and branch-wise comparisons validate its structural advantages. Buoy observations show peripheral errors below 3 m/s and physically consistent wind speed gradients in the core region. These results indicate that multi-source fusion of CYGNSS and environmental data, coupled with contrastive learning and physical reference, offers a reliable and efficient solution for typhoon wind speed retrieval. Full article

Mosque architecture often exhibits distinct identities, elements, and forms associated with geographical locations or dynastic patronage in the Islamic world. However, there has been a significant paradigm shift in mosque architecture over the past century, with external factors influencing the construction and sustainability of contemporary mosques. This study examines the evolution of mosque architecture in Nigeria, concentrating on the Ilorin Central Mosque as a pivotal case study connecting the northern and southern regions. The study employs a qualitative research methodology, utilizing descriptive approach, historical research, architectural analysis, and field observations to examine the architectural language, urban context, and socio-historical factors shaping the mosque’s development. Although geographical settings have always influenced traditional religious designs in Nigeria, the findings reveal a transformation from simple mud structures to grand modern edifices. The Ilorin Central Mosque exemplifies this shift, with its Ottoman-inspired domes and minarets contrasting with the traditional vernacular mosques of the 19th century. The study highlights the challenges of globalization, sustainability, foreign architectural influences, and the tension between local identity and contemporary trends in mosque architecture. The study concludes by arguing that future mosques must reintegrate regionalism, local materials, and climate-responsive principles into contemporary aesthetics while considering the quintessential principles of the Prophet’s Mosque and the religious and social significance of mosques within the urban fabric. The Ilorin Central Mosque exemplifies a microcosm of the transformations in Nigerian mosque architecture, highlighting the necessity of a balanced approach that embraces both cultural heritage and contemporary needs. Full article

The global production of dimension stones, that is, natural stones that can be processed into blocks and used as building and decorative materials, has grown steadily since the second half of the twentieth century. The rise of global markets and trade has also contributed to a rapid increase in imports of natural stones from distant locations. The introduction of dimension stones sourced from other continents can contribute significantly to geocultural heritage, defined as geological features that have acquired cultural, historical or symbolic meaning, as well as cultural elements embedded in a geological context. In the present contribution, the use of dimension stones in the city of Poznań (Poland, central Europe) is quantified. The study reveals dramatic changes in natural stone use between 1990 and 2019, with the number of dimension stone types increasing nearly threefold, and the mean distance to the stone source areas rising from 322 to 3885 km. Growing numbers and more diversified lithologies of natural stones can improve the urban landscape and contribute to the development of geotourism. On the other hand, increasing imports of dimension stones negatively affect local producers, threaten future conservation efforts, and have significant geoethical implications. Full article

In recent years, extracting coastlines from high-resolution remote sensing imagery has proven difficult due to complex details and variable targets. Current methods struggle with the fact that CNNs cannot model long-range dependencies, while Transformers incur high computational costs. To address these issues, we propose CVMFusion: a land–sea segmentation network based on a U-shaped encoder–decoder structure, whereby both the encoder and decoder are hierarchically organized. This architecture integrates the local feature extraction capabilities of CNNs with the global interaction efficiency of Mamba. The encoder uses parallel ConvNeXtV2 and VMamba branches to capture fine-grained details and long-range context, respectively. This network incorporates Dynamic Multi-Scale Attention (DyMSA) and Dynamic Weighted Cross-Attention (DyWCA) modules, which replace the traditional concatenation with an adaptive fusion mechanism to effectively fuse the features from the dual-branch encoder and utilize skip connections to complete the fusion between the encoder and decoder. Experiments on two public datasets demonstrate that CVMFusion attained MIoU accuracies of 98.05% and 96.28%, outperforming existing methods. It performs particularly well in segmenting small objects and intricate boundary regions. Full article

Underwater optical images are the primary carriers of underwater scene information, playing a crucial role in marine resource exploration, underwater environmental monitoring, and engineering inspection. However, wavelength-dependent absorption and scattering severely deteriorate underwater images, leading to reduced contrast, chromatic distortions, and loss of structural details. To address these issues, we propose a U-shaped underwater image enhancement framework that integrates Swin-Transformer blocks with lightweight attention and residual modules. A Dual-Window Multi-Head Self-Attention (DWMSA) in the bottleneck models long-range context while preserving fine local structure. A Global-Aware Attention Map (GAMP) adaptively re-weights channels and spatial locations to focus on severely degraded regions. A Feature-Augmentation Residual Network (FARN) stabilizes deep training and emphasizes texture and color fidelity. Trained with a combination of Charbonnier, perceptual, and edge losses, our method achieves state-of-the-art results in PSNR and SSIM, the lowest LPIPS, and improvements in UIQM and UCIQE on the UFO-120 and EUVP datasets, with average metrics of PSNR 29.5 dB, SSIM 0.94, LPIPS 0.17, UIQM 3.62, and UCIQE 0.59. Qualitative results show reduced color cast, restored contrast, and sharper details. Code, weights, and evaluation scripts will be released to support reproducibility. Full article

With rapid advancements in sub-meter satellite and aerial imaging technologies, high-resolution remote sensing imagery has become a pivotal source for geospatial information acquisition. However, current semantic segmentation models encounter two primary challenges: (1) the inherent trade-off between capturing long-range global context and preserving precise local structural details—where excessive reliance on downsampled deep semantics often results in blurred boundaries and the loss of small objects and (2) the difficulty in modeling complex scenes with extreme scale variations, where objects of the same category exhibit drastically different morphological features. To address these issues, this paper introduces MAFMamba, a multi-scale adaptive fusion visual Mamba network tailored for high-resolution remote sensing images. To mitigate scale variation, we design a lightweight hybrid encoder incorporating an Adaptive Multi-scale Mamba Block (AMMB) in each stage. Driven by a Multi-scale Adaptive Fusion (MSAF) mechanism, the AMMB dynamically generates pixel-level weights to recalibrate cross-level features, establishing a robust multi-scale representation. Simultaneously, to strictly balance local details and global semantics, we introduce a Global–Local Feature Enhancement Mamba (GLMamba) in the decoder. This module synergistically integrates local fine-grained features extracted by convolutions with global long-range dependencies modeled by the Visual State Space (VSS) layer. Furthermore, we propose a Multi-Scale Cross-Attention Fusion (MSCAF) module to bridge the semantic gap between the encoder’s shallow details and the decoder’s high-level semantics via an efficient cross-attention mechanism. Extensive experiments on the ISPRS Potsdam and Vaihingen datasets demonstrate that MAFMamba surpasses state-of-the-art Convolutional Neural Network (CNN), Transformer, and Mamba-based methods in terms of mIoU and mF1 scores. Notably, it achieves superior accuracy while maintaining linear computational complexity and low memory usage, underscoring its efficiency in complex remote sensing scenarios. Full article

Retrieval-Augmented Generation (RAG) is widely used for long-text summarization due to its efficiency and scalability. However, standard RAG methods flatten documents into independent chunks, disrupting sequential flow and thematic structure, resulting in significant loss of contextual information. This paper presents MOEGAT, a novel graph-enhanced retrieval framework that addresses this limitation by explicitly modeling document structure. MOEGAT constructs an Orthogonal Context Graph to capture sequential discourse and global semantic relationships—long-range dependencies between non-adjacent text spans that reflect topical similarity and logical associations beyond local context. It then employs a query-aware Mixture-of-Experts Graph Attention Network to dynamically activate specialized reasoning pathways. Experiments conducted on three public long-text summarization datasets demonstrate that MOEGAT achieves state-of-the-art performance. Notably, on the WCEP dataset, it outperforms the previous state-of-the-art Graph of Records (GOR) baseline by 14.9%, 18.1%, and 18.4% on ROUGE-L, ROUGE-1, and ROUGE-2, respectively. These substantial gains, especially the 14.9% improvement in ROUGE-L, reflect significantly better capture of long-range coherence and thematic integrity in summaries. Ablation studies confirm the effectiveness of the orthogonal graph and Mixture-of-Experts components. Overall, this work introduces a novel structure-aware approach to RAG that explicitly models and leverages document structure through an orthogonal graph representation and query-aware Mixture-of-Experts reasoning. Full article

To address the challenges in unstructured orchard environments, including high geometric similarity between fruit trees (with the measured average Euclidean distance difference between point cloud descriptors of adjacent trees being less than 0.5 m), significant dynamic interference (e.g., interference from pedestrians or moving equipment can occur every 5 min), and uneven terrain, this paper proposes an improved mapping algorithm named OSC-LIO (Orchard Scan Context Lidar Inertial Odometry via Smoothing and Mapping). The algorithm designs a dynamic point filtering strategy based on Euclidean clustering and spatiotemporal consistency within a 5-frame sliding window to reduce the interference of dynamic objects in point cloud registration. By integrating local semantic features such as fruit tree trunk diameter and canopy height difference, a two-tier verification mechanism combining “global and local information” is constructed to enhance the distinctiveness and robustness of loop closure detection. Motion compensation is achieved by fusing data from an Inertial Measurement Unit (IMU) and a wheel odometer to correct point cloud distortion. A three-level hierarchical indexing structure—”path partitioning, time window, KD-Tree (K-Dimension Tree)”—is built to reduce the time required for loop closure retrieval and improve the system’s real-time performance. Experimental results show that the improved OSC-LIO system reduces the Absolute Trajectory Error (ATE) by approximately 23.5% compared to the original LIO-SAM (Tightly coupled Lidar Inertial Odometry via Smoothing and Mapping) in a simulated orchard environment, while enabling stable and reliable path planning and autonomous navigation. This study provides a high-precision, lightweight technical solution for autonomous navigation in orchard scenarios. Full article

Forest resources are among the most important ecosystems on the earth. The semantic segmentation and accurate positioning of ground objects in forest remote sensing (RS) imagery are crucial to the emergency treatment of forest natural disasters, especially forest fires. Currently, most existing methods for image semantic segmentation are built upon convolutional neural networks (CNNs). Nevertheless, these techniques face difficulties in directly accessing global contextual information and accurately detecting geometric transformations within the image’s target regions. This limitation stems from the inherent locality of convolution operations, which are restricted to processing data structured in Euclidean space and confined to square-shaped regions. Inspired by the graph convolution network (GCN) with robust capabilities in processing irregular and complex targets, as well as Swin Transformers renowned for exceptional global context modeling, we present a hybrid semantic segmentation framework for forest RS imagery termed GSwin–Unet. This framework embeds the GCN model into Swin–Unet architecture to address the issue of low semantic segmentation accuracy of RS imagery in forest scenarios, which is caused by the complex texture features, diverse shapes, and unclear boundaries of land objects. GSwin–Unet features a parallel dual-encoder architecture of GCN and Swin Transformer. First, we integrate the Zero-DCE (Zero-Reference Deep Curve Estimation) algorithm into GSwin–Unet to enhance forest RS image feature representation. Second, a feature aggregation module (FAM) is proposed to bridge the dual encoders by fusing GCN-derived local aggregated features with Swin Transformer-extracted features. Our study demonstrates that, compared with the baseline models TransUnet, Swin–Unet, Unet, and DeepLab V3+, the GSwin–Unet achieves improvements of 7.07%, 5.12%, 8.94%, and 2.69% in the mean Intersection over Union (MIoU) and 3.19%, 1.72%, 4.3%, and 3.69% in the average F1 score (Ave.F1), respectively, on the RGB forest RS dataset. On the NIRGB forest RS dataset, the improvements in MIoU are 5.75%, 3.38%, 6.79%, and 2.44%, and the improvements in Ave.F1 are 4.02%, 2.38%, 4.72%, and 1.67%, respectively. Meanwhile, GSwin–Unet shows excellent adaptability on the selected GID dataset with high forest coverage, where the MIoU and Ave.F1 reach 72.92% and 84.3%, respectively. Full article

The paper introduces ‘transformative’ curriculum initiatives such as education for sustainable development (ESD) and global citizenship education (GCED), which address ‘macro’ challenges such as climate change, together with ‘holistic’ approaches to student learning such as ‘social and emotional learning’ (SEL) and education for ‘life skills’, ‘21st century skills’, ‘transversal competencies’, AI-related ethics, and ‘health and well-being.’ These are reflected in Section 6 of the 2023 UNESCO Recommendation on Education for Peace, Human Rights and Sustainable Development. It is suggested that such broad goals put forward at global policy level may serve as inspiration for national context-specific programming, while also needing better integration of national insights and cultural differences into global discourse. The paper aims to provide insights to education policy-makers responsible for national curriculum, textbooks and other learning resources, teacher training and examination processes, helping them to promote the human values, integrity and sense of agency needed by students in a time of multiple global and personal challenges. This requires an innovative approach to curricula for established school subjects and can be included in curricula being developed for AI literacy and related ethics. Research into the integration of transformative and holistic dimensions into curricula, materials, teacher preparation, and assessment is needed, as well as ongoing monitoring and feedback. AI-supported networking and resource sharing at local, national and international level can support implementation of transformative and holistic learning, to maintain and strengthen the human dimensions of learning. Full article

Images captured underwater frequently suffer from color casts, blurring, and distortion, which are mainly attributable to the unique optical characteristics of water. Although conventional UIE methods rooted in physics are available, their effectiveness is often constrained, particularly in challenging aquatic and illumination conditions. More recently, deep learning has become a leading paradigm for UIE, recognized for its superior performance and operational efficiency. This paper proposes UCA-Net, a lightweight CNN-Transformer hybrid network. It incorporates multiple attention mechanisms and utilizes composite attention to effectively enhance textures, reduce blur, and correct color. A novel adaptive sparse self-attention module is introduced to jointly restore global color consistency and fine local details. The model employs a U-shaped encoder–decoder architecture with three-stage up- and down-sampling, facilitating multi-scale feature extraction and global context fusion for high-quality enhancement. Experimental results on multiple public datasets demonstrate UCA-Net’s superior performance, achieving a PSNR of 24.75 dB and an SSIM of 0.89 on the UIEB dataset, while maintaining an extremely low computational cost with only 1.44M parameters. Its effectiveness is further validated by improvements in various downstream image tasks. UCA-Net achieves an optimal balance between performance and efficiency, offering a robust and practical solution for underwater vision applications. Full article

Understanding how transport conditions and the built environment shape urban CO₂ emissions is critical for low-carbon urban development. This study analyses CO₂ emission intensity across fifty major Chinese cities using integrated ODIAC emissions, VIIRS night-time lights, traffic performance indicators, built-environment morphology, population/POI structure, and socioeconomic controls. We develop a GeoAI workflow that couples XGBoost modelling with SHAP interpretation, congestion-based city grouping, and 1 km grid-level GNNWR to map intra-urban spatial non-stationarity. The global model identifies night-time light intensity as the strongest predictor, followed by population density and building density. SHAP results reveal pronounced nonlinearities, with high sensitivity at low–medium levels and diminishing marginal effects as activity and density increase. Although transport indicators are less influential in the aggregate model, their roles differ across congestion regimes: in low-congestion cities, emissions align more consistently with overall activity intensity, whereas in high-congestion cities they respond more strongly to population distribution, motorisation, and built-form intensity, with less stable relationships. Grid-level GNNWR further shows that key mechanisms are spatially uneven within cities, with local effects concentrating in specific cores and corridors or fragmenting across multiple subareas. These findings demonstrate that emission drivers are context-dependent across and within cities. Accordingly, uncongested cities may gain more from activity-related energy-efficiency measures, while highly congested cities may require congestion-sensitive land-use planning, spatial-structure optimisation, and motorisation control. Integrating explainable GeoAI with regime differentiation and spatial heterogeneity mapping provides actionable evidence for targeted low-carbon planning. Full article

To address challenges such as small target size, high density, severe occlusion, complex background interference, and edge device computational constraints, a lightweight model, GAFNet, is proposed based on YOLO11n, optimized for rice pest detection in field environments. To improve feature perception, we propose the Global Attention Fusion and Spatial Pyramid Pooling (GAM-SPP) module, which captures global context and aggregates multi-scale features. Building on this, we introduce the C3-Efficient Feature Selection Attention (C3-EFSA) module, which refines feature representation by combining depthwise separable convolutions (DWConv) with lightweight channel attention to enhance background discrimination. The model’s detection head, Enhanced Ghost Detect (EGDetect), integrates Enhanced Ghost Convolution (EGConv), Squeeze-and-Excitation (SE), and Sigmoid-Weighted Linear Unit (SiLU) activation, which reduces redundancy. Additionally, we propose the Focal-Enhanced Complete-IoU (FECIoU) loss function, incorporating stability and hard-sample weighting for improved localization. Compared to YOLO11n, GAFNet improves Precision, Recall, and mean Average Precision (mAP) by 3.5%, 4.2%, and 1.6%, respectively, while reducing parameters and computation by 5% and 21%. GAFNet can deploy on edge devices, providing farmers with instant pest alerts. Further, GAFNet is evaluated on the AgroPest-12 dataset, demonstrating enhanced generalization and robustness across diverse pest detection scenarios. Overall, GAFNet provides an efficient, reliable, and sustainable solution for early pest detection, precision pesticide application, and eco-friendly pest control, advancing the future of smart agriculture. Full article