Search Results (3,103)

China’s inland regions are vital for territorial spatial planning and sustainable development due to their abundant resources. However, the dynamic coordination between natural resource utilization (NRU) and ecological resilience (ER) remains poorly understood. Using panel data from 20 inland provinces in China (2009–2023), this study constructs NRU and ER evaluation systems, with ER assessed through the Pressure–State–Response (PSR) framework. Indicator weights are determined using an AHP–entropy method. Kernel density, panel vector autoregression (P-VAR), and coupling coordination models are applied to examine spatiotemporal evolution patterns, coordination levels, and interaction mechanisms between NRU and ER. The results show that: (1) The NRU index rises overall, peaking around 2020 (0.706), while the intensity of resource development continues to decline. Regional disparities widen, resulting in a spatial pattern of development intensity that was higher in the west and lower in the east. (2) The ER index continues to rise, accelerating at certain stages, and reaches a peak (0.723) between 2018 and 2020. Geographically, the eastern region led the way, with values decreasing in a stepwise manner, and regional disparities showed relatively gradual changes. (3) The degree of coordination between the two continues to improve, evolving from a “low level of dispersion” to a “medium-to-high level of concentration.” This has resulted in a pattern where the eastern region leads, followed by the central and southwestern regions in succession. Specifically, the EC index rose from 0.429 to 0.615, and the CC index rose from 0.384 to 0.533. Eastern and Central China have already reached a medium level of coordination, while Northwest and Southwest China remain primarily at a basic level of coordination. (4) Significant bidirectional dynamic interactions exist between the NRU and ER, with asymmetric pathways. By region, the NE, EC, and NC exhibit greater fluctuations and higher system sensitivity, while the CC experiences more concentrated short-term shocks; the SW and NW exhibit relatively smoother responses and converge more rapidly. Policy implications highlight the need for region-specific coordination strategies, better alignment between resource development and ecological protection, and enhanced cross-regional governance to support sustainable inland development. Full article

Background: Solar energy provides a sustainable and environmentally friendly alternative to fossil fuels, and photovoltaic (PV) systems are increasingly deployed worldwide. However, their operational reliability is often compromised by various fault conditions, which reduce power output and shorten system lifespan. Although automated image-based deep learning methods have shown promise for PV fault classification, their performance is often limited by severe class imbalance and subtle, low-contrast defect patterns. This study aims to address these challenges by proposing an improved deep learning framework for robust PV fault classification. Method: An attention-enhanced convolutional neural network framework, termed AEConvNeXt, is proposed for PV fault classification. The model is built on a ConvNeXt-Tiny backbone and incorporates a dropout-regularized Convolutional Block Attention Module (CBAM) to enhance localized feature refinement. To further improve learning under imbalanced data conditions, a hybrid loss function combining Cross-Entropy Loss and Focal Loss is employed. Results: Experimental evaluations demonstrate that AEConvNeXt achieves an overall accuracy of 94.37% and a macro F1-score of 94.43%, outperforming the strongest baseline model, ResNet-50, by more than 3%. Grad-CAM visualizations further confirm that the model effectively focuses on fault-relevant regions, improving interpretability. The proposed framework also shows consistent and robust performance across all six PV fault categories under varying conditions. Conclusions: The proposed AEConvNeXt framework provides an accurate and explainable solution for real-time PV fault detection, effectively addressing class imbalance and improving minority fault recognition. Full article

Background/Objectives: The aim of this study was to develop and validate a robust, radiomics-based classification model that uses pre-treatment MRI to non-invasively differentiate among major soft tissue sarcoma (STS) subtypes and a benign mimic. Methods: In this retrospective study, a cohort of 332 patients with biopsy-proven leiomyosarcoma, myxofibrosarcoma, myxoid liposarcoma, dedifferentiated liposarcoma, and undifferentiated pleomorphic sarcoma, along with the benign mimic intramuscular myxoma, was analyzed. Pre-treatment T1-weighted fat-saturated contrast-enhanced and T2-weighted fat-saturated MRI sequences were used for analysis. Following manual tumor segmentation, 1240 three-dimensional radiomic features were extracted. An XGBoost classifier was trained and validated using a robust 250-iteration bootstrap framework with nested cross-validation to ensure rigorous feature selection and unbiased performance evaluation. The model’s performance was assessed independently on T1-only, T2-only, and combined T1+T2 feature sets. Results: The combined T1 and T2 model achieved superior performance with an accuracy of 0.68 ± 0.04 and an AUC of 0.92 ± 0.02. At the subtype level, balanced accuracy was highest for intramuscular myxoma (0.91 ± 0.05), dedifferentiated liposarcoma (0.84 ± 0.06), and leiomyosarcoma (0.83 ± 0.05). SHAP analysis identified key features driving predictions, such as low T2 GLSZM Zone Size Entropy for myxoma and high T2 GLSZM Gray-Level Variance for leiomyosarcoma, which aligns with known pathological characteristics. Misclassifications predominantly occurred between subtypes with overlapping radiomic profiles. Conclusions: Radiomics applied to pre-treatment MRI enables robust, non-invasive classification of STS subtypes, demonstrating strong clinical potential for improving diagnostic confidence and informing triage strategies. Full article

To address the inherent limitations of single-modality approaches in fine-grained bird species recognition, this paper proposes an adaptive audiovisual fusion method based on prediction confidence. The proposed framework comprises three core components: an image classification branch, an audio classification branch, and a confidence–adaptive fusion module. The image branch employs EfficientNet-B3 to extract fine-grained visual features through compound scaling and squeeze-and-excitation (SE) attention. The audio branch utilizes ResNet-50 to classify Mel spectrograms converted from bird vocalizations, incorporating a dense sampling inference strategy to fully exploit complete audio information. For multimodal integration, a confidence–adaptive fusion strategy is introduced that jointly considers information entropy and probability gap to dynamically assess the reliability of each modality’s prediction, thereby assigning fusion weights at the sample level without any additional trainable parameters. Experiments on the SSW60 multimodal bird recognition dataset show that the image branch achieves a Top-1 accuracy of 91.55%, outperforming ResNet-50 (89.75%) and VGG-16 (83.81%); the audio branch reaches 68.20%, surpassing AST (63.29%) and VGG-16 (53.48%); and the fused model attains 95.30% Top-1 accuracy, a 3.75 percentage-point improvement over the image-only baseline and a 0.21 percentage-point gain over the learning-based TMC fusion baseline without introducing any trainable parameters, confirming the effectiveness of the proposed method. Full article

Secure real-time communication is a critical requirement in modern electric vehicle (EV) networks. These networks transmit safety-critical control commands through vulnerable in-vehicle communication channels. This study proposes a novel three-dimensional symmetric chaotic system for high-security EV communication. The system exhibits extensive multistability and symmetric double-wing attractors. To enhance dynamical complexity, its parameters are optimized using chaotic-enhanced particle swarm optimization (C-PSO). The largest Lyapunov exponent is used as the optimization objective. A fixed-time nonlinear controller is designed for rapid drive–response synchronization. The settling-time bound is independent of the initial conditions. The proposed method is evaluated through realistic Controller Area Network (CAN) bus simulations. These simulations include 12-bit quantization and a 1 ms sampling period. The experimental results show synchronization within 0.057 s. The recovered signal achieves an MSE of $1.202\times {10}^{-4}$. The encrypted signal reaches a Shannon entropy of 7.9904. These results confirm accurate recovery, strong randomness, and improved resistance to cryptographic attacks. Full article

Accurate spatiotemporal forecasting of sea surface temperature (SST) makes a great difference to offshore wind power development, since SST is a crucial factor influencing wind field patterns. In this work, a remote sensing-driven, multi-parameter field-coupled co-forecasting approach is proposed to utilize the cross-field interaction mechanisms among different physical fields to enhance forecasting performance. With this approach, more than one physical field can be simultaneously forecasted, thus improving forecasting efficiency. Compared with pure SST forecasting cases, the advanced enhancement of SST forecasting performance based on this approach is achieved by coupling SST with geostrophic current (GC) in data-driven forecasting. Also, both the spatiotemporal SST and GC fields are demonstrated to be accurately forecasted simultaneously. In addition, the causal effects between SST and GC are demonstrated as a reliable factor for evaluating the coupling scheme. To further improve co-forecasting performance, an exponential cross-entropy loss function is proposed for multi-physical field co-forecasting scenes, and shows more satisfying performance than a classical cross-entropy loss function. The results demonstrate that the data-driven multi-physical field-coupled co-forecasting approach is an advanced, highly efficient method that can accurately forecast more than one physical field at the same time. Full article

To address peak-period congestion in high-density air route networks and the high cost and limited precision of traditional global control methods, this study proposes a congestion mitigation method based on pinning control theory. First, a comprehensive evaluation index system for critical waypoints is constructed from complex-network structural characteristics, traffic flow characteristics, and congestion-state information. Pearson correlation analysis is used to examine redundancy among candidate indicators, and the entropy-weighted TOPSIS method is then employed to evaluate waypoint importance and identify critical pinning nodes. Second, a GA-PID pinning control optimization model is established to realize closed-loop optimization of network congestion by dynamically regulating a small number of critical nodes. Finally, simulation experiments are conducted using actual operational trajectory data from the Yangtze River Delta airspace. The results show that the proposed method reduces the network congestion coefficient from 176 to 137, representing a decrease of 22.16%, and increases airspace resource utilization from 70.76% to 84.41%, representing an improvement of 19.29%. Compared with the baseline GA method, the proposed method achieves better optimization performance and requires adjustments at only 13 waypoints, whereas the baseline GA method requires adjustments at 25 waypoints, demonstrating lower control costs and higher regulation efficiency. Full article

Underwater images often exhibit low contrast and loss of detail due to light scattering and absorption, which poses significant challenges for visual analysis in aquatic environments. Polarization imaging addresses these issues by exploiting the polarization states of light, effectively reducing backscatter and enhancing image contrast. In this paper, we propose a polarization image fusion method guided by information entropy and a hierarchical-adaptive fusion strategy. Local information entropy is first employed to perform multiscale denoising on Degree of Linear Polarization (DOLP) images, enabling adaptive detail reconstruction while distinguishing texture from noise. Subsequently, a hierarchical fusion framework is applied: low-frequency components are enhanced through detail injection, while high-frequency components are fused using a structure-guided mechanism that leverages low-frequency gradient information to generate soft masks for phase-aligned detail integration and edge sharpening. Experiments conducted on self-collected underwater images, two public underwater datasets, and three general-scene datasets demonstrate that the proposed method improves objective metrics, including information entropy, average gradient, and edge strength. Subjective evaluations further confirm its effectiveness in preserving details and adapting to diverse scenes. Furthermore, rigorous ablation studies and runtime analyses demonstrate that the optimized framework achieves a highly favorable balance between robust, artifact-free detail enhancement and computational efficiency. The proposed approach provides a practical solution for underwater image enhancement, with potential applications in target detection and infrastructure inspection. Full article

Background: Human papillomavirus (HPV)-positive oropharyngeal squamous cell carcinoma (OPSCC) differs from HPV-negative OPSCC in its molecular, biological, and clinical characteristics. We reviewed the literature on radiological differences between HPV-positive and HPV-negative OPSCC across ultrasound, CT, MRI, and PET-CT as this appears to be a critical gap in the literature. Methods: We performed a narrative review of studies reporting imaging findings in OPSCC by HPV status. Two reviewers independently searched PubMed, Ovid MEDLINE, Ovid EMBASE, and the Cochrane Library, from inception to February 2026, supplemented by searching major radiology journals and the grey literature. Eligible English-language studies included patients with OPSCC of known HPV status, assessed at least one imaging modality, and reported imaging findings stratified by HPV status. After title, abstract, and full-text screening, 66 studies were included. Results: HPV(+) OPSCC was more commonly associated with well-defined primary tumours and a higher prevalence of nodal metastases, particularly cystic nodal metastases and extranodal extension. These findings were broadly concordant with reported radiomic signatures. MRI studies suggested lower apparent diffusion coefficient values, while PET-CT studies suggested higher entropy and smaller primary lesions in HPV-positive disease. Conclusions: Selected imaging features may help distinguish HPV(+) from HPV(−) OPSCC, but current evidence remains insufficient for reliable standalone clinical application. Prospective multicentre validation and integration of multimodal imaging, radiomics, and radiogenomics are needed to improve non-invasive HPV stratification and support future precision diagnostics. Full article

Intelligent mechanical fault diagnosis plays a key role in maintaining rotating machinery. Although data-driven unsupervised domain adaptation methods have achieved considerable progress, their industrial applications are often restricted by low-quality sensor data. Non-stationary vibration signals and background noise easily corrupt target pseudo-labels, while conventional methods focusing on global statistical matching usually neglect local structures, leading to confirmation bias under dynamic loads. To improve diagnostic reliability, we propose a Noise-Resilient Whitened Domain Adaptation (NRWDA) framework. To handle covariance fluctuations caused by changing working conditions, a Lipschitz-bounded Temporal Whitening (LTW) module is designed as a low-pass filter. An Entropy-guided Prototype Truncation (EPT) mechanism is adopted to discard ambiguous labels and better calibrate semantic centers. In addition, a Dispersion-Adaptive Contrastive Sharpening (DACS) strategy is introduced to dynamically adjust the contrastive temperature based on predictive dispersion, thus tightening decision boundaries. The proposed method is evaluated on CWRU, PU, and MFPT datasets. The PU dataset, featuring fluctuating loads and non-stationary signals, poses a strict test, yet our model maintains its stability even at a 0 dB SNR—a condition where standard approaches usually break down. During the $P0\to P3$ transfer task involving substantial radial force variations, NRWDA secures a 72.36% accuracy and surpasses established baselines. These findings confirm that our technique successfully isolates dependable diagnostic features from corrupted sensor measurements within actual industrial settings. Full article

Thunderstorms are the most frequent type of severe convective weather, which pose great threats to buildings, power transmission, communication facilities, and air transportation. Their analysis and forecasting have long been challenges in meteorological operations. Currently, deep learning-based lightning forecasting has a short valid period, mostly relying on satellite imagery, radar echoes, and lightning location data, focusing on very-short-range forecasting. The longest valid period does not exceed 6 h, and the forecasting accuracy is not high. Based on the physical quantities of the ECMWF numerical prediction model and the actual observations of single-point thunderstorms, this paper constructs a single-point thunderstorm forecasting model with a long validity period (>6 h). The study integrates multi-dimensional parameters such as thermal, dynamic, water vapor, and stratification instability, introduces the second-order moist potential vorticity S as a comprehensive predictor, systematically compares the forecasting performance of eight models, such as 1D PreRNN and ConvLSTM, and verifies the actual operational capability of the model through independent cases. The results show that the 1D PreRNN model has the best overall performance in all periods, which can effectively capture the temporal evolution characteristics of meteorological physical quantities and still has stable generalization performance under unbalanced samples. The model performs well in the 1st, 2nd, and 4th periods, and especially still has significant operational reference value in the 4th period with the longest forecasting validity period; only the 3rd period is weakly affected by the small number of samples. The effect of second-order moist potential vorticity has significant time-dependent differences. Its overall improvement effect is limited in short-term forecasting, but it can provide key disturbance signals in the 4th period with the longest forecasting validity period, and the model forecasting performance drops significantly after removal. The original binary cross-entropy loss is most suitable for the unbalanced sample scenario in this study, and weighted losses are prone to overcorrection. The method in this paper can achieve stable and reliable single-point thunderstorm forecasting for more than 6 h, and can provide long-term fixed-point meteorological support for key scenarios such as aerospace and new energy stations. Full article

In developing countries, promoting sustainable income growth for farmers is a major priority. This study constructs an evaluation index system for the whole agricultural industry chain from the perspective of synergy among the innovation chain, supply chain, value chain, and capital chain. It also empirically tests the enabling mechanisms and spatial effects of the whole agricultural industry chain on farmers’ income. The entropy value method was used to measure the development level of the whole agricultural industry chain. Two-way fixed effects, mediation effects, and spatial Durbin models were applied to investigate the impacts, mechanisms, and spatial characteristics of the whole agricultural industry chain on farmers’ income. The whole agricultural industry chain significantly promotes farmers’ income growth, with the expansion of the non-agricultural employment scale and the improvement of urbanization levels serving as the main pathways through which the whole agricultural industry chain drives increases in farmers’ income. The heterogeneity analysis reveals that the innovation chain and capital chain contribute the most prominent marginal effects; the effect intensity of the whole agricultural industry chain on farmers’ income presents a spatial gradient pattern of “Central > Western > Eastern”; and its income-increasing effect is more noticeable for middle- and low-income farmers, demonstrating significant pro-poor characteristics. Further analysis indicates that the whole agricultural industry chain exerts a significant positive spatial spillover effect on farmers’ income. Therefore, it is essential to optimize the layout of the whole agricultural industry chain, smooth the transmission channels of non-agricultural employment and urbanization, and enhance the benefit linkage mechanism targeting middle- and low-income farmers. Full article

The advancement of the rural digital economy (RDE) has created new opportunities to alleviate resource and environmental constraints in agriculture and to accelerate the green transformation of agricultural production. However, the existing literature provides limited analysis of the underlying mechanisms and insufficient identification of the multidimensional pathways through which the RDE promotes agricultural green total factor productivity (AGTFP). Using provincial panel data from 2013 to 2022, this study measures the level of the RDE with the entropy method and calculates AGTFP using the SBM-GML model. It further employs a two-way fixed effects model, a mediation model, and a threshold model to examine the specific mechanisms through which the RDE affects AGTFP. The results show that a 1% increase in the RDE is associated with a 3.7% increase in AGTFP, with significant positive effects on its decomposed indices. These findings remain robust after a series of robustness and endogeneity tests. Further analysis indicates that the RDE enhances AGTFP through improvements in rural human capital and agricultural-scale operations, with a significant threshold effect demonstrating increasing marginal returns beyond the threshold. The mediating effects are more pronounced in major grain-producing regions. Policy implications emphasize integrating the RDE with green agricultural production, promoting digital talent development and moderate-scale operations, and reducing regional disparities. These findings provide empirical evidence for fully leveraging the dividends of the RDE and advancing agricultural modernization. Full article

Comprehensive land consolidation (CLC) is a systematic initiative aimed at optimizing spatial patterns of land use and revitalizing idle rural land resources. It is a pivotal policy instrument for enhancing rural resilience and possesses significant practical implications. Grounded in resilience theory, this study establishes an evaluation system for rural resilience, assesses resilience levels in Huzhou from 2003 to 2023, and investigates its spatiotemporal characteristics employing the entropy-weighted TOPSIS method and geodetector model. Furthermore, this research identifies the driving factors and dynamic mechanisms through which comprehensive land consolidation impacts rural resilience. The study area is categorized into four zones based on land use types to elucidate regional heterogeneity. The findings indicate that comprehensive land consolidation markedly enhances rural resilience, which progresses from slow initial growth to accelerated improvement, ultimately culminating in leapfrog development. Spatially, rural resilience exhibits a “central-high, marginal-low” distribution, characterized by core-periphery agglomeration. Notably, the key driving factors vary significantly across different regions. Mechanistically, comprehensive land consolidation bolsters rural resilience through a sequential pathway that begins with consolidation intervention, which activates critical factors. This activation leads to structural reorganization within the rural framework, followed by the optimization of functions that enhance overall resilience. In terms of policy implications, it is essential to adopt differentiated consolidation strategies tailored to regional resource endowments, emphasizing the optimization of production-living-ecological spaces to foster integrated and sustainable rural development. Full article

This study evaluates the efficiency of fiscal support policies for village-level collective economies in S Province, a frontier region of China, over the analytical period of 2018–2023, which includes the policy implementation years (2019–2022) plus one pre-policy and one post-policy year. Integrating theories of collaborative governance, resource alertness, and inclusive rural development, we construct an efficiency measurement framework to assess policy performance across 13 regions. Static efficiency is measured using DEA-BCC and super-efficiency SE-DEA models, while dynamic total factor productivity (TFP) is analyzed via the DEA–Malmquist index. The entropy-weighted method is employed to ensure robust indicator weighting. The findings reveal the following: (1) The average super-efficiency is 0.855, indicating relatively high expenditure efficiency but significant regional disparities and room for improvement. (2) The TFP declined by an average of 9.7% over the analytical period (2018–2023), primarily due to technological regression, despite stable technical efficiency. Based on the TFP performance, regions are categorized into high-, middle-, and low-efficiency tiers. Accordingly, we propose policy recommendations including efficiency-driven funding allocation, long-term support mechanisms combining technological innovation and management empowerment, regionally differentiated strategies, and strengthened multi-stakeholder collaboration. This study provides empirical evidence for optimizing fiscal policies to promote the sustainable development of rural collective economies and advance inclusive rural development in frontier regions. Full article