Search Results (150)

Long-tailed image classification remains challenging for vision–language models. Head classes dominate training while tail classes are underrepresented and noisy, and short prompts with weak text supervision further amplify head bias. This paper presents TASA, an end-to-end framework that stabilizes textual supervision and enhances cross-modal fusion. A Semantic Distribution Modulation (SDM) module constructs class-specific text prototypes by cosine-weighted fusion of multiple LLM-generated descriptions with a canonical template, providing stable and diverse semantic anchors without training text parameters. Dual-Space Cross-Modal Fusion (DCF) module incorporates selective-scan state–space blocks into both image and text branches, enabling bidirectional conditioning and efficient feature fusion through a lightweight multilayer perceptron. Together with a margin-aware alignment loss, TASA aligns images with class prototypes for classification without requiring paired image–text data or per-class prompt tuning. Experiments on CIFAR-10/100-LT, ImageNet-LT, and Places-LT demonstrate consistent improvements across many-, medium-, and few-shot groups. Ablation studies confirm that DCF yields the largest single-module gain, while SDM and DCF combined provide the most robust and balanced performance. These results highlight the effectiveness of integrating text-driven prototypes with state–space fusion for long-tailed classification. Full article

In this study, a cross-hierarchical intelligent modeling framework integrating an ecological semantic encoder, a distribution-aligned contrastive loss, and a disturbance-aware attention mechanism was developed to address the semantic alignment challenge between aboveground vegetation and belowground seed banks within forest ecosystems. The proposed framework leverages artificial intelligence and deep learning to characterize the structural and functional coupling between vegetation and soil communities, thereby elucidating the ecological mechanisms that underlie forest regeneration and stability. Experiments using representative forest ecological plot datasets demonstrated that the model achieved a top-1 accuracy of 78.6%, a top-5 accuracy of 89.3%, a mean cosine similarity of 0.784, and a reduced Kullback–Leibler divergence of 0.128, while the Jaccard index increased to 0.512—surpassing traditional statistical and machine-learning baselines such as RDA, CCA, Procrustes, Siamese, and SimCLR. The model also reduced NMDS stress to 0.094 and improved the Sørensen coefficient to 0.713, reflecting high robustness and precision in reconstructing community structure and ecological distributions. Additionally, the integration of distribution alignment and disturbance-aware mechanisms allows the model to capture dynamic vegetation–soil feedbacks across environmental gradients and disturbance regimes. This enables more accurate identification of regeneration potential, resilience thresholds, and restoration trajectories in degraded forests. Overall, the framework provides a novel theoretical foundation and a data-driven pathway for applying artificial intelligence to forest ecosystem monitoring, degradation diagnosis, and adaptive management for sustainable recovery. Full article

To address the challenge of monitoring the postharvest jasmine bloom stages during industrial tea scenting processes, this study proposes an efficient U-shaped Network (U-Net) model with frequency–spatial cross-attention (FSCA-EUNet) to resolve critical bottlenecks, including repetitive backgrounds and small interclass differences, caused by stacked jasmine flowers during factory production. High-resolution images of stacked jasmine flowers were first preprocessed and input into FSCA-EUNet, where the encoder extracted multi-scale spatial features and the FSCA module incorporated frequency-domain textures. The decoder then fused and refined these features, and the final classification layer output the predicted bloom stage for each image. The proposed model was designed as a “U-Net”-like structure to preserve multiscale details and employed a frequency–spatial cross-attention module to extract high-frequency texture features via a discrete cosine transform. Long-range dependencies were established by NonLocalBlook, located after the encoders in the model. Finally, a momentum-updated center loss function was introduced to constrain the feature space distribution and enhance intraclass compactness. According to the experimental results, the proposed model achieved the best metrics, including 95.52% precision, 95.42% recall, 95.40% F1-score, and 97.24% mean average precision, on our constructed dataset with only 878.851 K parameters and 15.445 G Floating Point Operations (FLOPs), and enabled real-time deployment at 22.33 FPS on Jetson Orin NX edge devices. The ablation experiments validated the improvements contributed by each module, which significantly improved the fine-grained classification capability of the proposed network. In conclusion, FSCA-EUNet effectively addresses the challenges of stacked flower backgrounds and subtle interclass differences, offering a lightweight yet accurate framework that enables real-time deployment for industrial jasmine tea scenting automation. Full article

With the global energy transition and decarbonization goals, tower-type solar thermal power generation is increasingly important for dispatchable clean energy due to its high efficiency, thermal storage capacity, and regulation performance. However, current research focuses on ideal conditions, ignoring real geographical constraints on heliostat layout and environmental impacts on receiver performance. More practical scene modeling and performance evaluation methods are urgently needed. To address these issues, we propose a heliostat field simulation algorithm based on heat loss mechanisms and real site characteristics. The algorithm includes optical performance evaluation (cosine efficiency, shading, truncation, atmospheric transmittance) and heat loss mechanisms (radiation, convection, conduction) for realistic net heat output estimation. Experiments revealed the following: (1) higher central towers improve optical efficiency by increasing solar elevation angle; (2) radiation losses dominate at high power and tower height, while convection losses dominate at low power and tower height. Using the Economic-Integrated Score (EIS) optimization algorithm, we achieved optimal tower and receiver configuration with 40.22% average improvement over other configurations (maximum 3.9× improvement). This provides a scientific design basis for improving tower-type solar thermal systems’ adaptability and economy in different geographical environments. Full article

In recent years, the integration of flexible resources into active distribution networks (ADNs) has been significantly enhanced. By coordinating a variety of such resources, the economic efficiency, operational security, and overall stability of ADNs can be improved. In this study, a bi-level planning model is proposed for active distribution networks. The upper-level model aims to minimize the annual comprehensive cost, while the lower-level model focuses on reducing network losses. To solve the upper-level problem, a hybrid whale optimization algorithm (HWOA) is developed. The algorithm integrates adaptive mutation based on Gaussian–Cauchy distributions, a nonlinear cosine-based control strategy, and a dual-population co-evolution mechanism. These enhancements allow HWOA to achieve faster convergence, higher accuracy, and stronger global search capabilities, thereby reducing the risk of falling into local optima. The lower-level problem is addressed using the interior point method due to its nonlinear and continuous nature. The proposed model and algorithm are validated through simulations on the IEEE 33-bus system. The results show that DG consumption increases by 88.77 MWh, network losses decrease by 6.8 MWh, and the total system cost is reduced by CNY 3.62 million over the entire project lifecycle. These improvements contribute to both the economic and operational performance of the ADN. Compared with the polar fox optimization algorithm (PFA), HWOA improves algorithmic efficiency by 18.92%, lowers network loss costs by 6.22%, and reduces the total system costs by 0.71%, demonstrating its superior effectiveness in solving complex bi-level optimization problems in active distribution networks. These findings not only demonstrate the technical efficiency of the proposed method but also contribute to the long-term goals of sustainable energy systems by improving renewable energy utilization, reducing operational losses, and supporting carbon reduction targets in active distribution networks. Full article

This paper presents a heuristic methodology for the optimal expansion of unbalanced three-phase distribution systems in rural areas, simultaneously addressing feeder routing and conductor sizing to minimize the total annualized cost—defined as the sum of investments in conductors and operational energy losses. The planning strategy explores two radial topological models: the Minimum Spanning Tree (MST) and the Steiner Tree (ST). The latter incorporates auxiliary nodes to reduce the total line length. For each topology, an initial conductor sizing is performed based on three-phase power flow calculations using Broyden’s method, capturing the unbalanced nature of the rural networks. These initial solutions are refined via four metaheuristic algorithms—the Chu–Beasley Genetic Algorithm (CBGA), Particle Swarm Optimization (PSO), the Sine–Cosine Algorithm (SCA), and the Grey Wolf Optimizer (GWO)—under a master–slave optimization framework. Numerical experiments on 15-, 25- and 50-node rural test systems show that the ST combined with GWO consistently achieves the lowest total costs—reducing expenditures by up to 70.63% compared to MST configurations—and exhibits superior robustness across all performance metrics, including best-, average-, and worst-case solutions, as well as standard deviation. Beyond its technical contributions, the proposed methodology supports the United Nations Sustainable Development Goals by promoting universal energy access (SDG 7), fostering cost-effective rural infrastructure (SDG 9), and contributing to reductions in urban–rural inequalities in electricity access (SDG 10). All simulations were implemented in MATLAB 2024a, demonstrating the practical viability and scalability of the method for planning rural distribution networks under unbalanced load conditions. Full article

Tropical cyclones (TCs) rank among the most destructive natural hazards globally, with core damaging potential originating from regions of intense wind shear and steep wind speed gradients within the eyewall and spiral rainbands. Accurately characterizing these fine-scale structural features is therefore critical for understanding TC intensity evolution, wind hazard distribution, and disaster mitigation. Recently, the deep learning-based downscaling methods have shown significant advantages in efficiently obtaining high-resolution wind field distributions. However, existing methods are mainly used to downscale general wind fields, and research on downscaling extreme wind field events remains limited. There are two main difficulties in downscaling TC wind fields. The first one is that high-quality datasets for TC wind fields are scarce; the other is that general deep learning frameworks lack the ability to capture the dynamic characteristics of TCs. Consequently, this study proposes a novel deep learning framework, CycloneWind, for downscaling TC surface wind fields: (1) a high-quality dataset is constructed by integrating Cyclobs satellite observations with ERA5 reanalysis data, incorporating auxiliary variables like low cloud cover, surface pressure, and top-of-atmosphere incident solar radiation; (2) we propose CycloneWind, a dynamically constrained Transformer-based architecture incorporating three wind field dynamical operators, along with a wind dynamics-constrained loss function formulated to enforce consistency in wind divergence and vorticity; (3) an Adaptive Dynamics-Guided Block (ADGB) is designed to explicitly encode TC rotational dynamics using wind shear detection and wind vortex diffusion operators; (4) Filtering Transformer Layers (FTLs) with high-frequency filtering operators are used for modeling wind field small-scale details. Experimental results demonstrate that CycloneWind successfully achieves an 8-fold spatial resolution reconstruction in TC regions. Compared to the best-performing baseline model, CycloneWind reduces the Root Mean Square Error (RMSE) for the U and V wind components by 9.6% and 4.9%, respectively. More significantly, it achieves substantial improvements of 23.0%, 22.6%, and 20.5% in key dynamical metrics such as divergence difference, vorticity difference, and direction cosine dissimilarity. Full article

Cloud imagery analysis from terrestrial observation points represents a fundamental capability within contemporary atmospheric monitoring infrastructure, serving essential functions in meteorological prediction, climatic surveillance, and hazard alert systems. However, traditional ground-based cloud image segmentation methods have fundamental limitations, particularly their inability to effectively model the graph structure and symmetry in cloud data. To address this, we propose G-CLIP, a ground-based cloud image segmentation method based on graph symmetry. G-CLIP synergistically integrates four innovative modules. First, the Prototype-Driven Asymmetric Attention (PDAA) module is designed to reduce complexity and enhance feature learning by leveraging permutation invariance and graph symmetry principles. Second, the Symmetry-Adaptive Graph Convolution Layer (SAGCL) is constructed, modeling pixels as graph nodes, using cosine similarity to build a sparse discriminative structure, and ensuring stability through symmetry and degree normalization. Third, the Multi-Scale Directional Edge Optimizer (MSDER) is developed to explicitly model complex symmetric relationships in cloud features from a graph theory perspective. Finally, the Uncertainty-Driven Loss Optimizer (UDLO) is proposed to dynamically adjust weights to address foreground–background imbalance and provide uncertainty quantification. Extensive experiments on four benchmark datasets demonstrate that our method achieves state-of-the-art performance across all evaluation metrics. Our work provides a novel theoretical framework and practical solution for applying graph neural networks (GNNs) to meteorology, particularly by integrating graph properties with uncertainty and leveraging symmetries from graph theory for complex spatial modeling. Full article

Remote sensing objects often exhibit significant scale variations, high aspect ratios, and diverse orientations. The anisotropic spatial distribution of such objects’ features leads to the conflict between feature representation and boundary regression caused by the coupling of different attribute parameters: previous detection methods based on square-kernel convolution lack the overall perception of large-scale or slender objects due to the limited receptive field; if the receptive field is simply expanded, although more context information can be captured to help object perception, a large amount of background noise will be introduced, resulting in inaccurate feature extraction of remote sensing objects. Additionally, the extracted features face issues of feature conflict and discontinuous loss during parameter regression. Existing methods often neglect the holistic optimization of these aspects. To address these challenges, this paper proposes SODE-Net as a systematic solution. Specifically, we first design a multi-scale fusion and spatially orthogonal convolution (MSSO) module in the backbone network. Its multiple shapes of receptive fields can naturally capture the long-range dependence of the object without introducing too much background noise, thereby extracting more accurate target features. Secondly, we design a multi-level decoupled detection head, which decouples target classification, bounding-box position regression and bounding-box angle regression into three subtasks, effectively avoiding the coupling problem in parameter regression. At the same time, the phase-continuous encoding module is used in the angle regression branch, which converts the periodic angle value into a continuous cosine value, thus ensuring the stability of the loss value. Extensive experiments demonstrate that, compared to existing detection networks, our method achieves superior performance on four widely used remote sensing object datasets: DOTAv1.0, HRSC2016, UCAS-AOD, and DIOR-R. Full article

This paper presents an efficient U-Net architecture featuring a modified Global Local Former Block (mGLFB) for simultaneous speech denoising and resolution reconstruction. Optimized for computational efficiency in the discrete cosine transform domain, the proposed architecture reduces model size by 13.5% compared to a standard GLFB-based U-Net, while maintaining comparable performance across multiple quality metrics. In addition to the mGLFB redesign, we introduce a perceptual loss that better captures high-frequency magnitude spectra, yielding notable gains in high-resolution recovery, especially in unvoiced speech segments. However, the mGLFB-based U-Net still shows limitations in retrieving spectral details with substantial energy in 4–6 kHz frequencies. Full article

Landslide susceptibility assessment constitutes a pivotal method of preventing and reducing losses caused by geological disasters. However, traditional models are often influenced by subjective grading factors, which can result in unscientific and inaccurate assessment outcomes. In this study, we thoroughly analyze various landslide causative factors, including geological, topographical, hydrological, and environmental components. A quantitative continuous model was employed, with methods such as frequency ratio (FR), cosine amplitude (CA), information value (IV), and certainty factor (CF) being applied in order to assess the landslide susceptibility of the Wanzhou coastline in the Three Gorges Reservoir area. The results were then compared with methods such as Bias-Standardised Information Value (BSIV), Support Vector Machine (SVM), Random Forest (RF), and Gradient Boosted Decision Tree (GBDT). This process led to the following key conclusions: (1) Most landslide susceptibility zones are predominantly banded and clustered on both sides of the Dewuidu River, particularly along the left bank of the Yangtze River from Dewuidu Town to Wanzhou City, as well as in the main urban area of Wanzhou. Clusters of the Yangtze River mainstem and surrounding towns characterize these areas. (2) The enhanced statistical analysis model shows a notable increase in sensitivity to landslides, achieving an Area Under the Curve (AUC) of 0.8878 for the IV model—an improvement of 0.0639 over the traditional BSIV model. This enhancement aligns closely with machine learning capabilities, and the spatial results obtained are more continuous. (3) By substituting manual grading with a quantitative continuous model, we achieve a balance between interpretability and computational efficiency. These findings lay a scientific foundation for the prevention and management of geological disasters in Wanzhou and provide valuable insights for comparable regions undertaking landslide susceptibility assessments. Full article

Accurate and early classification of retinal diseases such as diabetic retinopathy, cataract, and glaucoma is essential for preventing vision loss and improving clinical outcomes. Manual diagnosis from fundus images is often time-consuming and error-prone, motivating the development of automated solutions. This study proposes a deep learning-based classification model using a pretrained EfficientNetB3 architecture, fine-tuned on a publicly available Kaggle retinal image dataset. The model categorizes images into four classes: cataract, diabetic retinopathy, glaucoma, and healthy. Key enhancements include transfer learning, data augmentation, and optimization via the Adam optimizer with a cosine annealing scheduler. The proposed model achieved a classification accuracy of 95.12%, with a precision of 95.21%, recall of 94.88%, F1-score of 95.00%, Dice Score of 94.91%, Jaccard Index of 91.2%, and an MCC of 0.925. These results demonstrate the model’s robustness and potential to support automated retinal disease diagnosis in clinical settings. Full article

Palmprint recognition is becoming more and more common in the fields of security authentication, mobile payment, and crime detection. Aiming at the problem of small sample size and low recognition rate of palmprint, a small-sample palmprint recognition method based on image expansion and Dynamic Model-Agnostic Meta-Learning (DMAML) is proposed. In terms of data augmentation, a multi-connected conditional generative network is designed for generating palmprints; the network is trained using a gradient-penalized hybrid loss function and a dual time-scale update rule to help the model converge stably, and the trained network is used to generate an expanded dataset of palmprints. On this basis, the palmprint feature extraction network is designed considering the frequency domain and residual inspiration to extract the palmprint feature information. The DMAML training method of the network is investigated, which establishes a multistep loss list for query ensemble loss in the inner loop. It dynamically adjusts the learning rate of the outer loop by using a combination of gradient preheating and a cosine annealing strategy in the outer loop. The experimental results show that the palmprint dataset expansion method in this paper can effectively improve the training efficiency of the palmprint recognition model, evaluated on the Tongji dataset in an N-way K-shot setting, our proposed method achieves an accuracy of 94.62% ± 0.06% in the 5-way 1-shot task and 87.52% ± 0.29% in the 10-way 1-shot task, significantly outperforming ProtoNets (90.57% ± 0.65% and 81.15% ± 0.50%, respectively). Under the 5-way 1-shot condition, there was a 4.05% improvement, and under the 10-way 1-shot condition, there was a 6.37% improvement, demonstrating the effectiveness of our method. Full article

During traditional aircraft landings, pilots face significant challenges in identifying runway numbers with the naked eye, particularly at decision height under adverse weather conditions. To address this issue, this study proposes a novel detection algorithm based on an optimized version of the YOLOv5s model (You Only Look Once, version 5) for recognizing runway markings during civil aircraft landings. By integrating a data augmentation strategy with external datasets, the method effectively reduces both false detections and missed targets through expanded feature representation. An Alpha Complete Intersection over Union (CIOU) Loss function is introduced in place of the original CIOU Loss function, offering improved gradient optimization. Additionally, the model incorporates several advanced modules and techniques, including a Convolutional Block Attention Module (CBAM), Soft Non-Maximum Suppression (Soft-NMS), cosine annealing learning rate scheduling, the FReLU activation function, and deformable convolutions into the backbone and neck of the YOLOv5 architecture. To further enhance detection, a specialized small-target detection layer is added to the head of the network and the resolution of feature maps is improved. These enhancements enable better feature extraction and more accurate identification of smaller targets. As a result, the optimized model shows significantly improved recall (R) and precision (P). Experimental results, visualized using custom-developed software, demonstrate that the proposed optimized YOLOv5s model achieved increases of 5.66% in P, 2.99% in R, and 2.74% in mean average precision (mAP) compared to the baseline model. This study provides valuable data and a theoretical foundation to support the accurate visual identification of runway numbers and other reference markings during aircraft landings. Full article

This paper presents a hybrid optimization framework for solving the optimal reactive power compensation problem in medium-voltage smart distribution networks. Leveraging Julia’s computational environment, the proposed method combines the global search capabilities of the Chu & Beasley genetic algorithm (CBGA) with the local refinement efficiency of the interior-point optimizer (IPOPT). The objective is to minimize the annualized operating costs by reducing active power losses while considering the investment and operating costs associated with thyristor-switched capacitors (TSCs). A key contribution of this work is the comparative assessment of fixed and time-varying reactive power injection strategies. Simulation results on the IEEE 33- and 69-bus test feeders demonstrate that the proposed CBGA-IPOPT framework achieves annualized cost reductions of up to 11.22% and 12.58% (respectively) under fixed injection conditions. With variable injection, cost savings increase to 12.43% and 14.08%. A time-domain analysis confirms improved voltage regulation, substation reactive demand reductions exceeding 500 kvar, and peak loss reductions of up to 32% compared to the uncompensated case. Benchmarking shows that the hybrid framework not only consistently outperforms state-of-the-art metaheuristics (the sine-cosine algorithm, the particle swarm optimizer, the black widow optimizer, and the artificial hummingbird algorithm) in terms of solution quality but also demonstrates high solution repeatability across multiple runs, underscoring its robustness. The proposed method is directly applicable to real-world distribution systems, offering a scalable and cost-effective solution for reactive power planning in smart grids. Full article