Search Results (69)

In natural agricultural environments, plant disease monitoring faces significant challenges, including a highly uneven (long-tail) distribution of disease species, tiny scales of early-stage lesions, and complex, variable backgrounds. These factors hinder the ability of existing lightweight models to balance detection accuracy and computational efficiency. To address these issues, this paper proposes a detection scheme driven by the synergy of data distribution reshaping and model architecture optimization. At the data level, we propose the CALM-Aug augmentation strategy. Based on the statistical distribution characteristics of disease categories, this strategy utilizes object-level copy-paste logic to specifically compensate for the feature shortcomings of rare disease samples. It introduces a teacher-guided screening mechanism and employs accept–reject sampling to ensure the pathological consistency of the augmented samples, thereby alleviating the model’s inductive bias toward head categories. At the model architecture level, using YOLOv11 as the baseline, the YOLO11-ARL model adapted to agricultural scenarios is constructed. It enhances sensitivity to early point-like disease spots through Efficient Multi-Scale Convolutional Pyramids and lightweight decoupled detection heads. Furthermore, a Layer-wise Adaptive Feature-guided Distillation Pruning (LAFDP) algorithm is utilized to extract a lightweight version, YOLO11-ARL-PD, achieving a significant reduction in parameters and computational cost. Experimental results on the PlantDoc dataset show that the final model achieves a precision of 89.0% and an mAP@0.5 of 85.3%. Compared to the baseline model YOLOv11n, YOLO11-ARL-PD improves precision and average precision by 7.7 and 2.6 percentage points, respectively, while reducing parameters by 51.93% and weights by 46.15%. Cross-dataset tests prove the good generalization performance of the proposed method. This study indicates that, under lightweight constraints, jointly optimizing the training distribution and model architecture is an effective way to improve plant disease monitoring and to support the edge deployment of smart crop-protection systems. All resources for CALM-Aug are available at wyz-2004/CALM-Aug on GitHub. Full article

Olive cultivation represents a key pillar of rural economies and cultural heritage in Mediterranean regions, including western Greece. Despite its socio-economic importance, the sector faces increasing pressures from climate change, market volatility, and technological transformation, while progress toward environmentally sustainable production remains uneven. This study investigates how olive farmers’ perceptions of carbon footprint and climate risks are influenced by their demographic characteristics. Primary data were collected through 402 structured questionnaires distributed to olive producers in the Aetolia–Acarnania region. The sample was designed to represent farmers directly engaged in olive production, ensuring the relevance and reliability of the collected data. The findings, based on descriptive statistics, reveal significant heterogeneity in producers’ perceptions of climate risks and their capacity to respond through sustainable practices. Demographic characteristics appear to play an important role in shaping awareness of carbon footprint and the potential adoption of environmentally responsible farming strategies. These results suggest that sustainability transitions in perennial cropping systems depend not only on technological availability but also on social, informational, and institutional capacities. Strengthening agricultural advisory services, farmer training, and climate adaptation strategies may therefore support the adoption of climate-smart practices in olive cultivation. Furthermore, cooperation and value-chain integration are identified as potentially important mechanisms for facilitating knowledge transfer and supporting the adoption of sustainable practices (e.g., efficient irrigation and optimized input use). However, their contribution to environmental performance and greenhouse gas mitigation cannot be directly inferred from the present perception-based analysis and should be examined in future research using appropriate quantitative or environmental assessment frameworks. Full article

With the large-scale integration of renewable energy and the increasing use of power electronics, the issue of wide-band oscillations in power grids has become increasingly prominent. The scarcity and uneven distribution of oscillation samples pose significant challenges for training data-driven models, and traditional generative models struggle to ensure fidelity in both time and frequency domains. To address this, this paper proposes a Time–Frequency Fusion Generative Adversarial Network (TFF-GAN) for generating power grid oscillation risk scenarios. The method constructs a dual-path generation and discrimination framework, where the generator decomposes the signal using Short-Time Fourier Transform (STFT), with time-domain features extracted by a convolutional neural network (CNN) and frequency-domain features extracted from the STFT representation by a dedicated spectral network. These features are then fused using a U-Net structure. The discriminator simultaneously evaluates the authenticity of both the time-domain waveform and the frequency-domain spectrum. A composite loss function, incorporating time-domain loss, frequency-domain loss, and adversarial loss, is used for joint optimization. Experimental results demonstrate that the proposed method generates oscillation scenarios with high fidelity in both time-domain waveforms and frequency-domain spectra, effectively supporting power grid oscillation risk assessment and control strategy validation. Full article

Accurate estimation of solar radiation is of great significance for solar energy development and climate research. However, in China, the scarcity and uneven distribution of observation stations often cause deep learning models to overfit and suffer from accuracy degradation under small-sample conditions. To address this issue, this paper proposes a deep learning framework that integrates transfer learning and multi-scale time series modeling for predicting hourly global solar radiation at target meteorological sites. The method employs representation learning and clustering to select source domain sites with similar climatic characteristics. It integrates wavelet transform convolution, depthwise separable convolution, and a Transformer encoder–decoder to achieve multi-scale feature extraction and long-term dependency modeling. Experimental results demonstrate that the model achieved a coefficient of determination (R²) of 0.9710 in tests conducted in the Ningxia region. It maintained good predictive performance even in a cold-start scenario with only one month of training data and exhibited stable accuracy across all four seasons, effectively mitigating seasonal bias. This provides a reliable solution for solar radiation estimation in data-scarce regions, and its modeling approach can also be extended to other climate-related time series prediction tasks. Full article

The ceramic defect detection technology based on deep learning suffers from the problems of scarce rare defect samples and class imbalance. However, the current deep generative image augmentation techniques are limited when applied to the task of augmenting rare ceramic defects due to issues such as uneven image brightness and insufficient features of small-sized defects, resulting in poor image quality and limited improvement in detection results. This paper proposes a ceramic rare defect image augmentation method based on TC-CycleGAN. TC-CycleGAN is based on the CycleGAN framework and optimizes the generator and discriminator structures to make them more suitable for ceramic defect features, thereby improving the quality of generated images. The generator is TC-UNet, which introduces the scSE and DehazeFormer modules on the basis of UNet, effectively enhancing the model’s ability to learn the subtle defect features on the ceramic surface; the discriminator is the TC-PatchGAN architecture, which replaces the original BatchNorm module with the ContraNorm module, effectively increasing the discriminator’s sensitivity to the representation of tiny ceramic defect features and enhancing the diversity of generated images. The image quality assessment experiments show that the method proposed in this paper significantly improves the quality of generated defective images. For the concave type images, the FID and KID values have decreased by 49% and 73%, respectively, while for the smoke stains type images, the FID and KID values have decreased by 57% and 63% respectively. The further defect detection experiments results show that when using the data set expanded by the method in this paper for training, the recognition accuracy of the detection model for rare defects has significantly improved. The detection accuracy of the concave and smoke stains types of defects has increased by 1.2% and 3.9% respectively. Full article

We present a visual–inertial odometry (VIO) system that integrates a deep feature extraction and filtering strategy with optical flow to improve tracking robustness. While many traditional VIO methods rely on hand-crafted features, they often struggle to remain robust under challenging visual conditions, such as low texture, motion blur, or lighting variation. These methods tend to exhibit large performance variance across different environments, primarily due to the limited repeatability and adaptability of hand-crafted keypoints. In contrast, learning-based features offer richer representations and can generalize across diverse domains thanks to data-driven training. However, they often suffer from uneven spatial distribution and temporal instability, which can degrade tracking performance. To address these issues, we propose a hybrid front-end that combines a lightweight deep feature extractor with an image pyramid and grid-based keypoint sampling to enhance spatial diversity. Additionally, a forward–backward optical-flow-consistency check is applied to filter unstable keypoints. The system improves feature tracking stability by enforcing spatial and temporal consistency while maintaining real-time efficiency. Finally, the effectiveness of the proposed VIO system is validated on the EuRoC MAV benchmark, showing a 19.35% reduction in trajectory RMSE and improved consistency across multiple sequences compared with previous methods. Full article

Remote sensing scene image classification has extensive application scenarios in fields such as land use monitoring and environmental assessment. However, traditional methodologies based on convolutional neural networks (CNNs) face considerable challenges caused by uneven image quality, imbalanced sample distribution, intra-class similarities and limited computing resources. To address such issues, this study proposes a hierarchical lightweight attention-enhanced network (HLAE-Net), which employs a hierarchical feature collaborative extraction (HFCE) strategy. By considering the differences in resolution and receptive field as well as the varying effectiveness of attention mechanisms across different network layers, the network uses different attention modules to progressively extract features from the images. This approach forms a complementary and enhanced feature chain among different layers, forming an efficient collaboration between various attention modules. In addition, an improved lightweight attention module group is proposed, including a lightweight dual coordinate spatial attention module (DCSAM), which captures spatial and channel information, as well as the lightweight multiscale spatial and channel attention module. These improved modules are incorporated into the featured average sampling (FAS) bottleneck and basic bottlenecks. The experiments were studied on four public standard datasets, and the results show that the proposed model outperforms several mainstream models from recent years in overall accuracy (OA). Particularly in terms of small training ratios, the proposed model shows competitive performance. Maintaining the parameter scale, it possesses both good classification ability and computational efficiency, providing a strong solution for the task of image classification. Full article

Existing educational leadership research consistently emphasizes the importance of empowering and supporting classroom teachers to develop essential teaching experiences and leadership skills, enabling them to become autonomous curriculum developers and thinkers. This study aimed to explore the perceptions and understanding of distributed pedagogical leadership among Kenyan preservice professional actors in their respective contexts. It also examined the significance and impact of this practice on enhancing and strengthening the teaching and leadership abilities of teacher educators, thereby empowering them as effective pedagogical leaders in the classroom. The study employed a mixed methods design with a convergent parallel approach, using purposive sampling to select 83 participants, including administrative leaders, formal teacher leaders, and teacher educators from five public and private preservice teacher training colleges. Data collection involved semi-structured interviews, focus group discussions for qualitative insights, and an online survey for quantitative data. Results show that principals and formal teacher leaders play a key role in empowering teacher educators by distributing pedagogical leadership responsibilities among all professional actors. However, teacher educators felt that the distribution of tasks and responsibilities was uneven, which hindered effective implementation. This study also highlights how employer policies, through principals, influence the distribution of pedagogical leadership responsibilities. Full article

CT scanners are essential tools in modern medical imaging. Sudden failures of their X-ray tubes can lead to equipment downtime, affecting healthcare services and patient diagnosis. However, existing prediction methods based on a single model struggle to adapt to the multi-stage variation characteristics of tube lifespan and have limited modeling capabilities for temporal features. To address these issues, this paper proposes an intelligent prediction architecture for CT tubes’ remaining useful life based on a dual-branch neural network. This architecture consists of two specialized branches: a residual self-attention BiLSTM (RSA-BiLSTM) and a multi-layer dilation temporal convolutional network (D-TCN). The RSA-BiLSTM branch extracts multi-scale features and also enhances the long-term dependency modeling capability for temporal data. The D-TCN branch captures multi-scale temporal features through multi-layer dilated convolutions, effectively handling non-linear changes in the degradation phase. Furthermore, a dynamic phase detector is applied to integrate the prediction results from both branches. In terms of optimization strategy, a dynamically weighted triplet mixed loss function is designed to adjust the weight ratios of different prediction tasks, effectively solving the problems of sample imbalance and uneven prediction accuracy. Experimental results using leave-one-out cross-validation (LOOCV) on six different CT tube datasets show that the proposed method achieved significant advantages over five comparison models, with an average MSE of 2.92, MAE of 0.46, and R² of 0.77. The LOOCV strategy ensures robust evaluation by testing each tube dataset independently while training on the remaining five, providing reliable generalization assessment across different CT equipment. Ablation experiments further confirmed that the collaborative design of multiple components is significant for improving the accuracy of X-ray tubes remaining life prediction. Full article

In road scenes, the collected 3D point cloud data is usually accompanied by a large amount of interference mainly composed of ground point clouds and the property of uneven density distribution, which will bring difficulties to subsequent recognition and prediction. To address these problems, this paper proposes a ground point cloud segmentation algorithm based on local density plane fitting. Firstly, for the uneven density distribution of 3D point clouds, density segmentation is used to obtain several regions with balanced density. Then, candidate sample selection and plane validity detection are carried out for each region. The modified classical DBSCAN clustering algorithm is used to obtain effective fitting planes and perform clustering according to the fitting planes. Finally, different planes are divided according to the clustering results, and abnormal inspection is performed on the obtained results to screen out the most reasonable result. This scheme can effectively improve the scalability of the algorithm, reduce training costs, and improve deployment efficiency and universality. Experimental results show that the algorithm used in this paper has advantages compared with advanced algorithms of the same category, and can greatly reduce ground interference. Full article

The unique characteristics of desert vegetation, such as different leaf morphology, discrete canopy structures, sparse and uneven distribution, etc., pose significant challenges for remote sensing-based estimation of fractional vegetation cover (FVC). The Unmanned Aerial Vehicle (UAV) system can accurately distinguish vegetation patches, extract weak vegetation signals, and navigate through complex terrain, making it suitable for applications in small-scale FVC extraction. In this study, we selected the floodplain fan with Caragana korshinskii Kom as the constructive species in Hatengtaohai National Nature Reserve, Bayannur, Inner Mongolia, China, as our study area. We investigated the remote sensing extraction method of desert sparse vegetation cover by placing samples across three gradients: the top, middle, and edge of the fan. We then acquired UAV multispectral images; evaluated the applicability of various vegetation indices (VIs) using methods such as supervised classification, linear regression models, and machine learning; and explored the feasibility and stability of multiple machine learning models in this region. Our results indicate the following: (1) We discovered that the multispectral vegetation index is superior to the visible vegetation index and more suitable for FVC extraction in vegetation-sparse desert regions. (2) By comparing five machine learning regression models, it was found that the XGBoost and KNN models exhibited relatively lower estimation performance in the study area. The spatial distribution of plots appeared to influence the stability of the SVM model when estimating fractional vegetation cover (FVC). In contrast, the RF and LASSO models demonstrated robust stability across both training and testing datasets. Notably, the RF model achieved the best inversion performance (R² = 0.876, RMSE = 0.020, MAE = 0.016), indicating that RF is one of the most suitable models for retrieving FVC in naturally sparse desert vegetation. This study provides a valuable contribution to the limited existing research on remote sensing-based estimation of FVC and characterization of spatial heterogeneity in small-scale desert sparse vegetation ecosystems dominated by a single species. Full article

Terahertz metamaterial dual-band absorbers are used for multi-target detection and high-sensitivity sensing in complex environments by enhancing information that reflects differences in the measured substances. Traditional design processes are complex and time-consuming. Machine learning-based methods, such as neural networks and deep learning, require a large number of simulations to gather training samples. Existing design methods based on single-objective optimization often result in uneven multi-objective optimization, which restricts practical applications. In this study, we developed a metamaterial absorber featuring a circular split-ring resonator with four gaps nested in a “卍” structure and used the Multi-Objective Firefly Algorithm based on Multiple Cooperative Strategies to achieve fast optimization of the absorber’s structural parameters. A comparison revealed that our approach requires fewer iterations than the Multi-Objective Particle Swarm Optimization and reduces design time by nearly half. The absorber designed using this method exhibited two resonant peaks at 0.607 THz and 0.936 THz, with absorptivity exceeding 99%, indicating near-perfect absorption and quality factors of 31.42 and 30.08, respectively. Additionally, we validated the absorber’s wave-absorbing mechanism by applying impedance-matching theory. Finally, we elucidated the resonance-peak formation mechanism of the absorber based on the surface current and electric-field distribution at the resonance frequencies. These results confirmed that the proposed dual-band metamaterial absorber design is efficient, representing a significant step toward the development of metamaterial devices. Full article

Satellite image-based farmland classification plays an essential role in agricultural monitoring. However, typical tiling-based classification approaches, which extract patches at fixed offsets within each image during training, often suffer from structural issues such as patch duplication, limiting training diversity. Additionally, farmland classification frequently exhibits class imbalance due to uneven cultivation areas, resulting in biased training toward majority classes and poorer performance on minority classes. To overcome these issues, we propose Class-Balanced Random Patch Training, which combines Random Patch Extraction (RPE) and Class-Balanced Sampling (CBS). This method improves patch-level diversity and ensures balanced class representation during training. We evaluated our method on the FarmMap dataset, using a validation set from the same region and year as the training data, and a test set from a different year and region to simulate domain shifts. Our approach improved the F1 scores of minority classes and overall performance. Furthermore, our analysis across varying levels of class difficulty showed that the method consistently outperformed other configurations, regardless of minority-class difficulty. These results demonstrate that the proposed method offers a practical and generalizable solution for addressing class imbalance in tiling-based remote sensing classification, particularly under real-world conditions with spatial and temporal variability. Full article

The gearbox, as an important transmission component in wind turbines, connects the blades to the generator and is responsible for converting wind energy into mechanical energy and transmitting it to the generator. Its ability to reduce vibrations directly affects the operational lifespan of the wind turbine. When designing the gearbox’s elastic support structure, it is essential to evaluate how the design parameters influence various performance metrics. Neural networks offer a powerful means of capturing and interpreting the intricate associations linking structural parameters with performance metrics. However, conventional neural networks are usually optimized for a single task, failing to fully account for task differences and shared information. This can lead to task conflicts or insufficient feature modeling, which in turn affects the learning efficiency of inter-task correlations. Furthermore, physical experiments are costly and provide limited training, making it difficult to meet the large-scale dataset requirements for neural network training. To address the high cost and limited scalability of traditional physical testing for gearbox rubber damping structures, in this study, we propose a low-cost performance prediction method that replaces expensive experiments with simulation-driven dataset generation. An optimal Latin hypercube sampling technique is employed to generate high-quality data at minimal cost. On this basis, a multi-task prediction model called multi-gate mixture-of-experts with LSTM (PLE-LSTM) is constructed. The adaptive gating mechanism, hierarchical nonlinear transformation, and effective capture of temporal dynamics in the LSTM significantly enhance the model’s ability to model complex nonlinear patterns. During training, a dynamic weighting strategy named GradNorm is utilized to counteract issues like the early stabilization in multi-task loss convergence and the uneven minimization of loss values. Finally, ablation experiments conducted on different datasets validate the effectiveness of this approach, with experimental results demonstrating its success. Full article

Digital transformation is reshaping educational methods and necessitates new institutional, teaching, and learning approaches. In Romania’s secondary school system, this process is particularly complex due to infrastructural disparities, uneven digital competencies, and limited policy guidance. This study investigates the factors influencing Romanian secondary school teachers’ adoption of digital technologies, addressing a gap in the literature regarding localized teacher behavior within under-digitalized educational environments. Using the Technology Acceptance Model (TAM), this research explores perceived usefulness, perceived ease of use, behavioral intention, and perceived quality as determinants of digital adoption. The analysis was conducted on a sample of 430 teachers utilizing Partial Least Squares Structural Equation Modeling (PLS-SEM). The key findings reveal that behavioral intention strongly predicts actual usage (β = 0.791, p < 0.001), and perceived ease of use exerts a greater influence than perceived usefulness. Furthermore, perceived quality plays a modest but significant mediating role in enhancing digital engagement. These results suggest that user-friendly, intuitive technologies and targeted professional training are essential for sustained adoption. This study contributes to the field by offering a context-specific understanding of technology’s acceptance in Eastern European education and by extending the TAM framework through the integration of perceived quality. It provides actionable insights for policymakers and school leaders seeking to promote sustainable and inclusive digital transformation in schools. Full article