Search Results (11,214)

Utilizing multi-source meteorological data for lightning identification is crucial for monitoring severe convective weather. However, several key challenges persist in this field: dimensional imbalance and modal competition among multi-source heterogeneous data, model training bias caused by the extreme sparsity of lightning samples, and an imbalance between false alarms and missed detections resulting from complex background noise. To address these challenges, this paper proposes a lightning identification network guided by physical priors and constrained by supervision. First, to tackle the issue of modal competition in fusing satellite (high-dimensional) and radar (low-dimensional) data, a physical prior-guided asymmetric radar information enhancement mechanism is introduced. This mechanism uses radar physical features as contextual guidance to selectively enhance the latent weak radar signatures. Second, at the architectural level, a multi-source multi-scale feature fusion module and a weighted sliding window–multilayer perceptron (MLP) enhanced decoding unit are constructed. The former achieves the coupling of multi-scale physical features at a 2 km grid scale through cross-level semantic alignment, building a highly consistent feature field that effectively improves the model’s ability to detect lightning signals. The latter leverages adaptive receptive fields and the nonlinear modeling capability of MLPs to effectively smooth spatially discrete noise, ensuring spatial continuity in the reconstructed results. Finally, to address the model bias caused by severe class imbalance between positive and negative samples—resulting from the extreme sparsity of lightning events—an asymmetrically weighted BCE-DICE loss function is designed. Its “asymmetric” characteristic is implemented by assigning different penalty weights to false-positive and false-negative predictions. This loss function balances pixel-level accuracy and inter-class equilibrium while imposing high-weight penalties on false-positive predictions, achieving synergistic optimization of feature enhancement and directional suppression. Experimental results show that the proposed method effectively increases the hit rate while substantially reducing the false alarm rate, enabling efficient utilization of multi-source data and high-precision identification of lightning strike areas. Full article

RGBT tracking holds irreplaceable value in unmanned aerial vehicle (UAV) ground observation missions, effectively supporting scenarios such as nighttime monitoring and low-altitude reconnaissance. However, existing frameworks based on CNNs or Transformers face inherent trade-offs between interaction capabilities and computational efficiency. Furthermore, current methods perform poorly in challenging scenarios involving target scale variations and rapid motion from UAV perspectives. To address these issues, this paper proposes a novel multimodal interaction and fusion Mamba network (MIFMNet), which achieves fundamental innovations relative to existing RGB-T fusion trackers and recent Mamba-based tracking methods. Different from existing RGB-T trackers that rely on CNN’s local convolution or Transformer’s quadratic-complexity self-attention for cross-modal fusion, MIFMNet departs from these architectures and designs modality-adaptive interaction mechanisms based on Mamba, fully leveraging the complementary information while resolving the efficiency-accuracy trade-off. Specifically, this paper designs the scale differential enhanced Mamba (SDEM), which expands the receptive field through multiscale parallel convolutions while amplifying complementary information via differential strategies to enhance feature responses to scale-varying objects. Furthermore, we propose flow-guided multilayer interaction Mamba (FMIM), which integrates inter-frame motion information into scanning prediction. This enables the network to adaptively adjust interaction priorities between shallow texture and high-level semantic features based on motion intensity, mitigating early information forgetting and enhancing robustness in dynamic scenes. Extensive experiments on four major benchmarks demonstrate that MIFMNet achieves state-of-the-art performance on precision and success rate, particularly excelling in UAV scenarios involving occlusion, scale variations, and rapid motion. Simultaneously, it achieves an inference speed of 35.3 FPS, enabling efficient deployment on resource-constrained platforms, thereby providing robust support for UAV applications of RGBT tracking. Full article

The East China Sea represents a critical coastal wetland region, characterized by complex geomorphology, heterogeneous land-cover composition, and diverse wetland types. Accurate delineation of coastal wetland extent is essential for ecosystem service assessment and sustainable coastal management, directly contributing to wetland-related Sustainable Development Goals (SDGs), particularly SDG 15, on ecosystem conservation and biodiversity protection. However, pronounced spectral similarity and structural heterogeneity among wetland classes pose substantial challenges to reliable classification. To address these challenges, this study developed a hierarchical classification framework integrating Random Forest, K-means clustering, and a decision tree classifier based on multi-source Sentinel-1 and Sentinel-2 imagery. Spectral, polarimetric, texture, and morphological features were systematically constructed to enhance class separability. Using this framework, a 10 m resolution coastal wetland map of the East China Sea was generated for 2023. The proposed approach achieved an overall accuracy of 91.32% and improved the discrimination of spectrally similar wetland types. Feature fusion reduced confusion among water-related classes, while object-based clustering improved the extraction of linear riverine wetlands. The resulting 10 m wetland map provides updated spatial information for ecological assessment and coastal management in the East China Sea. Full article

Aiming to address the complex coupling of transient impulses and steady-state components in vibration signals of scroll compressors in air conditioning systems, this study proposes a physically driven heterogeneous dual-path adaptive weighting network (PDW-Net). The approach constructs a physics-inspired weighting module based on kurtosis and energy criteria, enabling adaptive reconstruction of transient impulses and steady-state vibration components. Feature extraction and decision-level fusion are achieved through a heterogeneous dual-branch network comprising a Fast Fourier Transform (FFT)-based one-dimensional convolutional neural network (1D-CNN) and a Short-Time Fourier Transform (STFT)-based two-dimensional convolutional neural network (2D-CNN). In experimental validation covering four typical fault conditions—condenser failure, refrigerant deficiency, refrigerant overcharge, and main shaft wear—the PDW-Net achieved an average diagnostic accuracy of 97.87% (standard deviation: 2.60%), with 100% accuracy in identifying refrigerant deficiency and normal operating states, demonstrating significant superiority over existing mainstream methods. Ablation studies reveal that the adaptive weighting mechanism contributes most substantially to performance, as its removal results in a 34.24 percentage point drop in accuracy. Replacing the heterogeneous dual-branch structure with a homogeneous counterpart reduces accuracy by 16.18 percentage points, robustly validating the efficacy of the physics-guided and heterogeneous fusion design. Full article

Satellite remote sensing plays an essential role in wildfire monitoring due to its large-scale observation capability. However, fire targets in satellite imagery are typically extremely small, sparsely distributed, and embedded in complex backgrounds, making accurate segmentation highly challenging for existing methods. To address these challenges, this paper proposes a multi-scale Mamba-based network for tiny fire segmentation, named FireMambaNet. The network adopts a nested U-shaped encoder-decoder architecture, primarily consisting of three modules: the Cross-layer Gated Residual U-shaped module (CG-RSU), the Fire-aware Directional Context Modulation module (FDCM), and the Multi-scale Mamba Attention Module (M2AM). The CG-RSU, as the core building block, adaptively suppresses background redundancy and enhances weak fire responses by extracting multi-scale features through cross-layer gating. The FDCM explicitly enhances the network’s ability to perceive anisotropic expansion features of fire points, such as those along the wind direction and terrain orientation, by modeling multi-directional context. The M2AM model employs a Mamba state-space model to suppress background interference through global context modeling during cross-scale feature fusion, while enhancing consistency among sparsely distributed tiny fire targets. In addition, experimental validation is conducted using two subsets from the Active Fire dataset, which have significant pixel-level sparse features: Oceania and Asia4. The results show that the proposed method significantly outperforms various mainstream CNN, Transformer, and Mamba baseline models on both datasets. It achieves an IoU of 88.51% and F1 score of 93.76% on the Oceania dataset, and an IoU of 85.65% and F1 score of 92.26% on the Asia4 dataset. Compared to the best-performing CNN baseline model, the IoU is improved by 1.81% and 2.07%, respectively. Overall, the FireMambaNet demonstrates significant advantages in detecting tiny fire points in complex backgrounds. Full article

In intelligent transportation systems and urban traffic management, accurate vehicle area intrusion detection based on surveillance imagery plays a critical role in ensuring road safety and operational efficiency. However, under real-world road surveillance conditions characterized by complex backgrounds, varying illumination, occlusion, and scale variations, mainstream detection algorithms often suffer from high false detection and missed detection rates, limiting their reliability and practical deployment. To address these challenges, this paper proposes YOLOv10-Intrusion, a high-precision vehicle area intrusion detection framework based on an improved version of YOLOv10s. The proposed algorithm incorporates Omni-Dimensional Dynamic Convolution (ODConv) and a custom-designed RCS_M module to enhance feature extraction and fine-grained recognition capability. In addition, a Bidirectional Feature Pyramid Network (BiFPN) is employed to optimize multi-scale feature fusion at the neck level. These improvements collectively reduce false detections and missed detections while improving model recall and mean Average Precision (mAP). Furthermore, the Wise-IoU (WIoU) loss function replaces the original Complete IoU (CIoU) loss to accelerate convergence and stabilize bounding box regression under complex surveillance conditions. A dedicated vehicle area intrusion dataset is constructed from real-world road surveillance footage, covering five vehicle categories across diverse road environments and lighting conditions. Experimental results demonstrate that, compared with the baseline YOLOv10s, YOLOv10-Intrusion achieves improvements of 1.5, 3.3, 3.6, and 2.8 percentage points in Precision, Recall, mAP@0.5, and mAP@0.5:0.95, respectively, and outperforms other mainstream detection algorithms in vehicle area intrusion detection tasks. Full article

Remote sensing (RS) object detection faces intrinsic challenges arising from the overhead imaging paradigm and the diversity of climatic conditions. In particular, atmospheric phenomena such as clouds and haze cause severe visual degradation, making reliable object detection difficult. However, most existing detectors are developed under clear-weather conditions, which limits their generalization capability in realistic haze-degraded RS scenarios. To alleviate this issue, an adaptive spectral-aware network for RS object detection under haze interference is proposed, termed RShDet, which is designed to handle both high-altitude RS imagery and low-altitude Unmanned Aerial Vehicle (UAV) scenarios. Firstly, the Object-Centered Dynamic Enhancement (OCDE) module dynamically adjusts the spatial positions of key-value pairs through query-agnostic offsets, enabling the network to emphasize object-relevant regions while suppressing haze-induced background interference. Secondly, the Dynamic Multi-Spectral Perception and Filtering (DSPF) module introduces a multi-spectral attention mechanism that adaptively selects informative frequency components, thereby enhancing discriminative feature representations in hazy environments. Thirdly, the Frequency-Domain Multi-Feature Fusion (FDMF) module employs learnable weights to complementarily integrate amplitude and phase information in the frequency domain, enabling effective cross-task feature interaction between the enhancement and detection branches. Extensive experiments demonstrate that RShDet consistently achieves superior detection performance under hazy conditions across both synthetic and real-world benchmarks. Specifically, it achieves improvements of 2.4% mAP50 on Hazy-DOTA, 1.9% mAP on HazyDet, and 2.33% mAP on the real-world foggy dataset RTTS, surpassing existing state-of-the-art methods. Full article

Traditional social recommendation methods often focus on static representations of users and items, neglecting dynamic changes in user interests and item attractiveness over time, which makes it challenging to adapt to temporal variations in user interests. Additionally, the propagation of information along explicit social relationships tends to over-smooth features and weaken individual preferences, while static implicit relationships may increase short-term noise. Thus, a Dual-channel Controllable Diffusion Network based on Hybrid Representations (HR-DCDN) is proposed for social recommendation. The HR-DCDN first incorporates temporal factors by combining dynamic and static representations to capture changes in user interests and item attractiveness. Then, our method proposes a dual-channel aggregation mechanism to obtain higher-order representations of users and items. Explicit social relationships serve as the social-influence channel, while implicit social relationships discovered via dynamic implicit relationship mining constitute the preference-homophily channel. In addition, a learnable polynomial spectral filter incorporates residual connections and dual-channel fusion information at each propagation step, stabilizing deep propagation and alleviating representation homogenization to a limited extent while preserving high-frequency preference information. Finally, we jointly optimize a cross-layer InfoNCE objective on the perturbed interaction branch with the supervised rating loss, which provides an additional empirical regularization effect, improves robustness, and helps preserve representation diversity without altering the graph structure. Experimental results demonstrate that our model outperforms baseline methods on two real-life social datasets. Full article

Synthetic Aperture Radar (SAR) enables all-weather maritime surveillance, yet ship-oriented bounding box (OBB) detection remains challenging in complex scenes. Strong sea clutter and dense harbor scatterers often mask the slender characteristics of ships as well as the weak responses of small ships. Meanwhile, the periodicity of angle parameterization introduces regression discontinuities, and near-symmetric, bright-scatterer-dominated signatures further cause heading ambiguity, undermining the stability of orientation prediction. Moreover, in most detectors, multi-scale feature fusion and angle estimation lack explicit coordination, and rotated-box localization performance is often jointly affected by feature degradation and unstable orientation prediction. To this end, we propose a unified framework that simultaneously strengthens multi-scale representations and stabilizes orientation modeling. Specifically, we design a Gaussian-Guided Stage-Aware Deformable Feature Pyramid Network (GSDFPN) and a Coarse-to-Fine Unit-Circle Resolver (CF-UCR). GSDFPN enhances multi-scale fusion with two plug-in components: (i) a Gaussian-guided High-level Semantic Refinement Module (GHSRM) that suppresses clutter-dominated semantics while strengthening ship-responsive cues, and (ii) a Stage-aware Deformable Fusion Module (SDFM) for low-level features, which disentangles channels into a geometry-preserving spatial stream and a clutter-resistant semantic stream, and couples them via deformable interaction with bidirectional cross-stream gating to better capture the inherent slender characteristics of ships and localize small ships. For orientation, CF-UCR decomposes angle prediction into direction-cluster classification and intra-cluster residual regression on the unit circle, effectively mitigating periodicity-induced discontinuities and stabilizing rotated-box estimation. On SSDD+ and RSDD, our method achieves AP/AP${}_{50}$/AP${}_{75}$ of 0.5390/0.9345/0.4529 and 0.4895/0.9210/0.4712, respectively, while reaching AP${}_{s75}$/AP${}_{m75}$/AP${}_{l75}$ of 0.5614/0.8300/0.8392 and 0.4986/0.8163/0.8934, evidencing strong rotated-box localization across target scales in complex maritime scenes. Full article

Defect detection in the hot rolling process is closely linked to the quality of the final product. Among these defects, slab camber during the intermediate rolling stage is one of the primary manifestations of asymmetry, which significantly impairs both the quality of the finished strip and the stability of subsequent rolling processes. Conventional image-based edge detection methods for slab camber are prone to detection deviations in complex industrial environments, mainly due to their weak noise robustness. To address the scientific challenge of low accuracy and poor robustness in feature extraction for hot-rolled intermediate slab camber detection, which is induced by environmental interference in complex industrial settings, we break through the technical bottlenecks of traditional edge detection methods and existing deep learning models in terms of channel–spatial feature collaborative optimization and anti-interference fusion of multi-scale features. We establish a dense perception network model integrated with a channel–spatial attention mechanism, realize robust feature recognition of slab edges under complex working conditions, and provide theoretical and technical support for the real-time quantitative detection of slab shape defects in the hot rolling process. The proposed model significantly improves detection accuracy and robustness through multi-scale feature enhancement and noise suppression, effectively meeting the requirements for real-time quantitative detection of slab camber in the roughing rolling stage. Field experiments verify that the method increases detection accuracy by 36.55% and achieves favorable performance on evaluation metrics, including ODS and OIS. Full article

This paper investigates how hybrid temples are created and transformed into tourist attractions, focusing on Wat Phothikyan Phutthathum—a Thai–Chinese temple located in Kelantan, close to Malaysia’s border with Thailand. This study aims to understand how both the local Thai minority and Chinese majority contribute to temple hybridisation, examine the influence of such temples on community identity, and explore their growing importance as tourist attractions. It highlights the negotiation and cultural exchange that shape new heritage spaces for borderland communities. Using visual analysis and interviews, the research shows that, since there are no Chinese temples nearby, Chinese Buddhists and Taoists adapt Thai temples by incorporating their own rituals and art. This results in blended iconography and practices, guided by an open-minded Thai monk. Features like large Buddha statues, staircases featuring naga-dragon designs, and murals combining different traditions reveal this fusion. The temple’s unique artwork and spiritual environment attract visitors from Muslim-majority areas and various countries like Thailand, Taiwan, and Singapore. As tourism becomes central to the temple’s role, the local authorities emphasise its religious significance and multicultural symbolism, aligning with economic interests and daily interactions among minority groups. This study offers valuable empirical and theoretical perspectives on the blending of religious heritage sites in border regions. Full article

Low-dose computed tomography (LDCT) can effectively reduce ionizing radiation; however, the associated image noise and artifacts can severely compromise the accuracy of clinical diagnosis. To address the challenge of balancing noise suppression and detail preservation in LDCT images, this study proposes a deep learning (DL)-based image denoising method termed Progressive Fusion Distillation Network (PFDN). Building upon the Information Multi-distillation Network (IMDN), the proposed method incorporates a pixel attention (PA) mechanism and a progressive fusion strategy, and further designs a Pixel Parallel Extraction Block (PPEB) together with a Progressive Fusion Distillation Block (PFDB) to fully exploit multi-scale and multi-channel features, thereby optimizing the image denoising network through efficient feature separation and re-fusion. In addition, by explicitly leveraging the noise characteristics specific to LDCT images, the method establishes an end-to-end training framework suitable for medical imaging. Experimental results demonstrate that PFDN not only effectively reduces image noise and artifacts, but also enhances overall image quality while preserving diagnostically relevant image structures under the adopted evaluation setting. Full article

Wind turbine blade surface defect detection remains challenging due to large variations in defect scales, blurred edge textures, and severe interference from complex backgrounds, which often lead to insufficient detection accuracy and high false and missed detection rates. To address these issues, this paper proposes an improved RTDETR-based detection framework, termed CEA-DETR, for wind turbine blade surface defect inspection. First, a Cross-Scale Multi-Edge feature Extraction (CSME) backbone is designed by integrating multi-scale pooling and edge-enhancement units with a dual-domain feature selection mechanism, enabling effective extraction of fine-grained texture and edge features across different scales. Second, an Efficient Multi-Scale Feature Fusion Network (EMSFFN) is constructed to facilitate deep cross-level feature interaction through adaptive weighted fusion and multi-scale convolutional structures, thereby enhancing the representation of multi-scale defects. Furthermore, an adaptive sparse self-attention mechanism is introduced to reconstruct the AIFI module, strengthening global dependency modeling and guiding the network to focus on critical defect regions under complex background conditions. Experimental results demonstrate that CEA-DETR achieves mAP50 and mAP50:95 of 89.4% and 68.9%, respectively, representing improvements of 3.1% and 6.5% over the RT-DETR-r18 baseline. Meanwhile, the proposed model reduces computational cost (GFLOPs) by 20.1% and parameter count by 8.1%. These advantages make CEA-DETR more suitable for deployment on resource-constrained unmanned aerial vehicles (UAVs), enabling efficient and real-time autonomous inspection of wind turbine blades. Full article

Grammatical error correction (GEC) for Chinese learner English is still dominated by sentence-level modeling, which limits discourse-level consistency and weakens adaptation to learner-specific error profiles. From an instructional perspective, these limitations also reduce the value of automated feedback as a basis for spaced-repetition instructional structures sensitive to individual differences. This study proposes a symmetry-aware dual-encoder architecture for context-aware GEC in Chinese learner English. A context encoder captures preceding-sentence information, while a source encoder integrates BERT-based semantic representations with Bi-GRU-based syntactic features for the current sentence. A gated decoder performs asymmetric fusion of local and contextual evidence. To better reflect corpus-level tendencies in Chinese learner English, a CLEC-informed augmentation strategy generates synthetic errors using empirical category frequencies as a coarse sampling prior. Experiments on CoNLL-2014, JFLEG, and CLEC show consistent improvements over strong neural baselines in F_0.5 and GLEU under the current desktop-oriented implementation setting. Nevertheless, the integration of BERT, dual encoders, and gated decoding introduces non-negligible computational overhead, and the present system is therefore better suited to desktop writing-support scenarios than to strict real-time or large-scale online deployment. The proposed framework thus provides a practical technical basis for personalized grammar feedback and for future spaced-repetition instructional designs in ESL writing support. Full article

Bone marrow cell classification and nucleus segmentation in microscopic images are fundamental tasks for computer-aided diagnosis of lymphocytic leukemia. However, bone marrow cells from different subtypes exhibit high morphological similarity, and structural information is often constrained under optical microscopic imaging, posing challenges for stable and effective feature representation. To address this issue, we propose MTLL (Multitask Model on Lymphocytic Leukemia), a novel multitask approach that performs cell classification and nucleus segmentation within a unified network to exploit their complementary information. The model constructs a hybrid backbone for shared feature representation based on a CNN-Transformer architecture, in which Fuse-MBConv modules are tightly integrated with multilayer multi-scale transformers to enable deep fusion of local texture and global semantic information. For the segmentation branch, we design an AM (Atrous Multilayer Perceptron) decoder that combines atrous spatial pyramid pooling with multilayer perceptrons to fuse multi-scale information and accurately delineate nucleus boundaries. The classification branch incorporates prior knowledge of cell nuclei structures to capture subtle variations in cellular morphology and texture, thereby enhancing the model’s ability to distinguish between leukemia subtypes. Experimental results demonstrate that the MTLL model significantly outperforms existing advanced single-task and multi-task models in both lymphocytic leukemia classification and cell nucleus segmentation. These results validate the effectiveness of the multi-task feature-sharing strategy for lymphocytic leukemia diagnosis using bone marrow microscopic images. Full article