Search Results (2,350)

Major crops worldwide are affected by various diseases yearly, leading to crop losses in different regions. The primary methods for addressing crop disease losses include manual inspection and chemical control. However, traditional manual inspection methods are time-consuming, labor-intensive, and require specialized knowledge. The preemptive use of chemicals also poses a risk of soil pollution, which may cause irreversible damage. With the advancement of computer hardware, photographic technology, and artificial intelligence, crop disease recognition methods based on spectral and red–green–blue (RGB) images not only recognize diseases without damaging the crops but also offer high accuracy and speed of recognition, essentially solving the problems associated with manual inspection and chemical control. This paper summarizes the research on disease recognition methods based on spectral and RGB images, with the literature spanning from 2020 through early 2025. Unlike previous surveys, this paper reviews recent advances involving emerging paradigms such as State Space Models (e.g., Mamba) and Generative AI in the context of crop disease recognition. In addition, it introduces public datasets and commonly used evaluation metrics for crop disease identification. Finally, the paper discusses potential issues and solutions encountered during research, including the use of diffusion models for data augmentation. Hopefully, this survey will help readers understand the current methods and effectiveness of crop disease detection, inspiring the development of more effective methods to assist farmers in identifying crop diseases. Full article

Hyperspectral anomaly detection (HAD) aims to identify pixels that significantly differ from the background without prior knowledge. While deep learning-based reconstruction methods have shown promise, they often suffer from limited feature representation, inefficient training cycles, and sensitivity to imbalanced data distributions. To address these challenges, this paper proposes a novel contrastive–transfer-synergized dual-stream transformer for hyperspectral anomaly detection (CTDST-HAD). The framework integrates contrastive learning and transfer learning within a dual-stream architecture, comprising a spatial stream and a spectral stream, which are pre-trained separately and synergistically fine-tuned. Specifically, the spatial stream leverages general visual and hyperspectral-view datasets with adaptive elastic weight consolidation (EWC) to mitigate catastrophic forgetting. The spectral stream employs a variational autoencoder (VAE) enhanced with the RossThick–LiSparseR (R-L) physical-kernel-driven model for spectrally realistic data augmentation. During fine-tuning, spatial and spectral features are fused for pixel-level anomaly detection, with focal loss addressing class imbalance. Extensive experiments on nine real hyperspectral datasets demonstrate that CTDST-HAD outperforms state-of-the-art methods in detection accuracy and efficiency, particularly in complex backgrounds, while maintaining competitive inference speed. Full article

Railway passenger flow forecasting plays a critical role in operational efficiency and resource allocation for transportation systems. However, existing deep learning approaches suffer significant performance degradation when facing rare but high-impact events, primarily due to sample scarcity and their inability to distinguish between routine patterns and disruption regimes. To address these challenges, this study introduces EA-ARIMA-Informer, an adaptive forecasting framework that integrates entropy-augmented ARIMA with Informer through an entropy-guided regime-switching mechanism. The passenger flow series is characterized through a multi-dimensional entropy space comprising four complementary measures: Sample Entropy quantifies local regularity and predictability, Permutation Entropy captures the complexity of ordinal dynamics, Transfer Entropy measures causal information flow from external events (holidays, weather) to passenger demand, and the Conditional Entropy Growth Factor (CEGF)—a novel metric introduced herein—detects regime transitions by tracking the rate of uncertainty change between consecutive time windows. These entropy indicators serve dual roles as feature inputs for representation learning and as state identifiers for segmenting the time series into stable and fluctuating regimes with distinct predictability properties. An adaptive dual-path architecture is then designed accordingly: EA-ARIMA handles low-entropy stable regimes where linear seasonality dominates, while EA-Informer processes high-entropy fluctuating regimes requiring nonlinear residual modeling, with CEGF-guided gating dynamically controlling component weights. Unlike conventional black-box gating mechanisms, this entropy-based switching provides physically interpretable signals that explain when and why different model components dominate the forecast. The framework is validated on a large-scale dataset covering nearly 300 Chinese cities over three years (2017–2019), encompassing normal operations, holiday peaks, and extreme weather disruptions. Experimental results demonstrate that EA-ARIMA-Informer achieves a MAPE of 4.39% for large-scale cities and 7.82% for data-scarce small cities (Tier-3), substantially outperforming standalone ARIMA, XGBoost, and Informer, which yield 15.95%, 13.75%, and 12.87%, respectively, for Tier-3 cities. Ablation studies confirm that both entropy-based feature augmentation and CEGF-guided regime switching contribute significantly to these performance gains, establishing a new paradigm for interpretable and adaptive forecasting in complex transportation systems. Full article

Newton–Raphson-Based Optimizer (NRBO) is a recently proposed metaheuristic that combines mathematical search rules with population-based optimization; however, it still suffers from an insufficient balance between global exploration and local exploitation, limited local refinement accuracy, and weak adaptability in complex optimization scenarios. To address these limitations, this paper proposes an Improved Newton–Raphson-Based Optimizer (INRBO), which enhances the original framework through a multi-strategy augmentation mechanism. Specifically, INRBO integrates three complementary strategies: (1) an adaptive differential operator with a linearly decaying scaling factor to dynamically regulate exploration and exploitation throughout the search process; (2) a quadratic interpolation strategy that exploits high-quality individuals to improve local search directionality and precision; and (3) an elitist population genetic strategy that preserves superior solution characteristics while maintaining population diversity and preventing premature convergence. The performance of INRBO is systematically evaluated on the CEC2017 benchmark suite under multiple dimensions and compared with several state-of-the-art metaheuristic algorithms. Experimental results demonstrate that INRBO achieves superior optimization accuracy, convergence efficiency, and robustness across unimodal, multimodal, hybrid, and composite functions, which is further confirmed by statistical significance tests. In addition, INRBO is applied to mobile robot path planning in grid-based environments of different scales, where it consistently generates shorter, smoother, and safer paths than competing algorithms. Overall, the proposed INRBO provides an effective and robust optimization framework for global continuous optimization problems and real-world engineering applications, demonstrating both strong theoretical value and practical applicability. Full article

Semantic segmentation of remote sensing images is crucial for geospatial applications but is severely hampered by the prohibitive cost of pixel-level annotations. Although semi-supervised learning (SSL) offers a solution by leveraging unlabeled data, its application to Vision Transformers (ViTs) often encounters overfitting and even training instability under extreme label scarcity. To tackle these challenges, we propose a Curriculum-based Self-supervised and Semi-supervised Pipeline (CSSP). The pipeline adopts a staged, easy-to-hard training strategy, commencing with in-domain pretraining for robust feature representation, followed by a carefully designed finetuning stage to prevent overfitting. The pipeline further integrates a novel Difficulty-Adaptive ClassMix (DA-ClassMix) augmentation that dynamically reinforces underperforming categories and a Progressive Intensity Adaptation (PIA) strategy that systematically escalates augmentation strength to maximize model generalization. Extensive evaluations on the Potsdam, Vaihingen, and Inria datasets demonstrate state-of-the-art performance. Notably, with only 1/32 of the labeled data on the Potsdam dataset, the CSSP reaches 82.16% mIoU, nearly matching the fully supervised result (82.24%). Furthermore, we extend the CSSP to a semi-supervised domain adaptation (SSDA) scenario, termed Cross-Domain CSSP (CDCSSP), which outperforms existing SSDA and unsupervised domain adaptation (UDA) methods. This work establishes a stable and highly effective framework for training ViT-based segmentation models with minimal annotation overhead. Full article

The Poweshiek skipperling Oarisma poweshiek (Parker, 1870) (Lepidoptera: Hesperiidae) is an endangered prairie obligate butterfly native to the north central United States and southern Canada. Conservation efforts for this species rely on ex situ approaches for population augmentation and reintroductions. As such, improving our understanding of the behaviors of Poweshiek skipperlings and maximizing their reproductive output in an ex situ setting are critical for the success of associated conservation initiatives. In this study, we examined the frequency of breeding behaviors exhibited by Poweshiek skipperlings in relation to various environmental factors: sunlight intensity (measured in lux), ambient temperature, and age. Sunlight intensity was a significant predictor of breeding behavior frequency, but we did not detect an effect of ambient temperature on breeding behavior. We also documented a generally negative relationship between age and breeding behavior frequency for both sexes. The results of our study underscore the importance of natural sunlight in encouraging breeding behaviors in an ex situ conservation environment. Ex situ observations also can help identify environmental conditions that promote high levels of Poweshiek skipperling activity, which could be used to optimize the timing of in situ population surveys. Full article

Water resources are of critical importance across all ecological, social, and economic realms. Accurate extraction of water bodies is of significance to estimate the spatial coverage of water resources and to mitigate water-related disasters. Single-modal remote sensing images are often insufficient for accurate water body extraction due to limitations in spectral information, weather conditions, and speckle noises. Furthermore, state-of-the-art deep learning models may be constrained by data extensibility, feature transferability, model scalability, and task producibility. This manuscript presents an integrated GeoAI framework that enhances foundation models for efficient water body extraction with multi-modal remote sensing images. The proposed framework consists of a data augmentation module tailored for optical and synthetic aperture radar (SAR) remote sensing images, as well as extraction modules augmented by three popular foundation models, namely SAM, SAMRS, and CROMA. Specifically, optical and SAR images are preprocessed and augmented independently, encoded through foundation model backbones, and subsequently decoded to generate water body segmentation masks under single-modal and multi-modal settings. Full article

While the Internal Model Principle (IMP) and Disturbance Observer (DOB) are fundamental to robust control, their systematic equivalence within a unified framework has received limited attention. IMP-based control achieves robustness through the structural inclusion of signal generators, whereas DOB-based methods rely on extended state representations for disturbance estimation. This paper bridges this gap by designing a state-space Reduced-Order Disturbance Observer (RODOB)-based controller that achieves systematic equivalence with an IMP-based transfer function controller. As a design example, an IMP-based controller is synthesized using a Linear Quadratic Regulator (LQR) for an augmented system in error space, with reference inputs directly integrated into the RODOB structure to eliminate the need for additional filters. Simulations and hardware experiments on a Brushless DC (BLDC) motor verify that both structures exhibit consistent control input and output characteristics, significantly outperforming conventional cascade and PID strategies. Numerical stability during digital implementation is ensured via partial fraction expansion. Furthermore, a method for estimating equivalent disturbances—encompassing both external loads and model uncertainties—is proposed by leveraging RODOB states. These findings suggest significant potential for future applications in fault diagnosis and real-time condition monitoring. Full article

Electroencephalography-based motor imagery (EEG-MI) classification is a cornerstone of Brain–Computer Interface (BCI) systems, enabling the identification of motor intentions by decoding neural patterns within EEG signals. However, conventional methods, predominantly reliant on convolutional neural networks (CNNs), are proficient at extracting local temporal features but struggle to capture long-range dependencies and global contextual information. To address this limitation, we propose a Triple-path Heterogeneous Feature Collaboration Network (TPHFC-Net), which synergistically integrates three distinct temporal modeling pathways: a multi-scale Temporal Convolutional Network (TCN) to capture fine-grained local dynamics, a Transformer branch to model global dependencies via multi-head self-attention, and a Long Short-Term Memory (LSTM) network to track sequential state evolution. These heterogeneous features are subsequently fused adaptively by a dynamic gating mechanism. In addition, the model’s robustness and discriminative power are further augmented by a lightweight front-end denoising diffusion model for enhanced noisy feature representation and a back-end prototype attention mechanism to bolster the inter-class separability of non-stationary EEG features. Extensive experiments on the BCI Competition IV-2a and IV-2b datasets validate the superiority of the proposed model, achieving mean classification accuracies of 82.45% and 89.49%, respectively, on the subject-dependent MI task and significantly outperforming existing mainstream baselines. Full article

The classification of ship targets in remote sensing images holds significant application value in fields such as marine monitoring and national defence. Although existing research has yielded considerable achievements in ship classification, current methods struggle to distinguish highly similar ship categories for fine-grained classification tasks due to a lack of targeted design. Specifically, they exhibit the following shortcomings: limited ability to extract locally discriminative features; inadequate fusion of features at high and low levels of representation granularity; and sensitivity of model performance to background noise. To address this issue, this paper proposes a fine-grained classification framework for ship targets in remote sensing images based on Multi-Granularity Attention Network (MA-Net), specifically designed to tackle the aforementioned three major challenges encountered in fine-grained classification tasks for ship targets in remote sensing. This framework first performs multi-level feature extraction through a backbone network, subsequently introducing an Adaptive Local Feature Attention (ALFA) module. This module employs dynamic overlapping region segmentation techniques to assist the network in learning spatial structural combinations, thereby optimising the representation of local features. Secondly, a Dynamic Multi-Granularity Feature Fusion (DMGFF) module is designed to dynamically fuse feature maps of varying representational granularities and select key attribute features. Finally, a Feature-Based Data Augmentation (FBDA) method is developed to effectively highlight target detail features, thereby enhancing feature expression capabilities. On the public FGSC-23 and FGSCR-42 datasets, MA-Net attains top-performing accuracies of 93.12% and 98.40%, surpassing the previous best methods and establishing a new state of the art for fine-grained classification of ship targets in remote sensing images. Full article

This article describes the methodology for adjusting and comparing eleven variants of Transformer architectures for the classification of skin lesions using images: five variants of Google’s Vision Transformer (ViT) and six variants of Microsoft’s Swin Transformer. We present the methodology used to achieve these results, which includes meticulous hyperparameter tuning and a robust data augmentation strategy to address the class imbalance problem. This approach allowed us to surpass the state of the art on the DermaMNIST dataset with respect to CNN-based models, and achieve very competitive results on the ISIC Challenge 2019 dataset with respect to Transformer-based models. In addition, we employed the CheferCAM method to provide visual explanations that identify the most influential image regions in the models’ predictions. Full article

Remote sensing image semantic segmentation faces substantial challenges in training and transferring models across images with varying resolutions. This issue can be effectively mitigated by continuously learning knowledge derived from new resolutions; however, this learning process is severely plagued by catastrophic forgetting. To address this problem, this paper proposes a novel continual learning framework termed Language-Guided Resolution-Continual Learning (i.e., LaGu-RCL), which alleviates catastrophic forgetting through two complementary strategies. On the one hand, a multi-resolution image augmentation pipeline is introduced to synthesize higher- and lower-resolution variants for each training batch, allowing the model to learn from images of diverse resolutions at every training step. On the other hand, a language-guided learning strategy is proposed to aggregate features of the same resolution while separating those of different resolutions. This ensures that the knowledge acquired from previously learned resolutions is not disrupted by that from unseen resolutions, thereby mitigating catastrophic forgetting. To validate the effectiveness of the proposed approach, we construct MR-ExcavSeg, a multi-resolution dataset covering several counties in Chongqing, and conduct comparative experiments between LaGu-RCL and several state-of-the-art continual learning baselines. Experimental results demonstrate that LaGu-RCL achieves significantly superior segmentation performance and continual learning capability, verifying its advantages. Full article

Self-compacting concrete (SCC) offers significant advantages in construction due to its superior workability; however, optimizing SCC mixture design remains challenging because of complex nonlinear material interactions and increasing sustainability requirements. This study proposes an integrated, sustainability-oriented computational framework that combines machine learning (ML), SHapley Additive exPlanations (SHAP), and multi-objective optimization to improve SCC mixture design. A large and heterogeneous publicly available global SCC dataset, originally compiled from 156 independent peer-reviewed studies and further enhanced through a structured three-stage data augmentation strategy, was used to develop robust predictive models for key fresh-state properties. An optimized XGBoost model demonstrated strong predictive accuracy and generalization capability, achieving coefficients of determination of $R^2=0.835$ for slump flow and $R^2=0.828$ for $T_{50}$ time, with reliable performance on independent industrial SCC datasets. SHAP-based interpretability analysis identified the water-to-binder ratio and superplasticizer dosage as the dominant factors governing fresh-state behavior, providing physically meaningful insights into mixture performance. A cradle-to-gate life cycle assessment was integrated within a multi-objective genetic algorithm to simultaneously minimize embodied CO${}_2$ emissions and material costs while satisfying workability constraints. The resulting Pareto-optimal mixtures achieved up to 3.9% reduction in embodied CO${}_2$ emissions compared to conventional SCC designs without compromising performance. External validation using independent industrial data confirms the practical reliability and transferability of the proposed framework. Overall, this study presents an interpretable and scalable AI-driven approach for the sustainable optimization of SCC mixture design. Full article

Artificial intelligence (AI) is rapidly reshaping gastrointestinal (GI) surgery by enhancing decision-making, intraoperative performance, and postoperative management. The integration of AI-driven systems is enabling more precise, data-informed, and personalized surgical interventions. This review provides a state-of-the-art overview of AI applications in GI surgery, organized into four key domains: surgical simulation, surgical computer vision, surgical data science, and surgical robot autonomy. A comprehensive narrative review of the literature was conducted, identifying relevant studies of technological developments in this field. In the domain of surgical simulation, AI enables virtual surgical planning and patient-specific digital twins for training and preoperative strategy. Surgical computer vision leverages AI to improve intraoperative scene understanding, anatomical segmentation, and workflow recognition. Surgical data science translates multimodal surgical data into predictive analytics and real-time decision support, enhancing safety and efficiency. Finally, surgical robot autonomy explores the progressive integration of AI for intelligent assistance and autonomous functions to augment human performance in minimally invasive and robotic procedures. Surgical AI has demonstrated significant potential across different domains, fostering precision, reproducibility, and personalization in GI surgery. Nevertheless, challenges remain in data quality, model generalizability, ethical governance, and clinical validation. Continued interdisciplinary collaboration will be crucial to translating AI from promising prototypes to routine, safe, and equitable surgical practice. Full article

Existing unsupervised anomaly detection methods suffer from insufficient parameter precision, poor robustness to noise, and limited generalization capability. To address these issues, this paper proposes an Adaptive Diffusion Adversarial Evolutionary Network (ADAEN) for unsupervised anomaly detection in tabular data. The proposed network employs an adaptive hierarchical feature evolution generator that captures multi-scale feature representations at different abstraction levels through learnable attribute encoding and a three-layer Transformer encoder, effectively mitigating the gradient vanishing problem and the difficulty of modeling complex feature relationships that are commonly observed in conventional generators. ADAEN incorporates a multi-scale adaptive diffusion-augmented discriminator, which preserves scale-specific features across different diffusion stages via cosine-scheduled adaptive noise injection, thereby endowing the discriminator with diffusion-stage awareness. Furthermore, ADAEN introduces a multi-scale robust adversarial gradient loss function that ensures training stability through a diffusion-step-conditional Wasserstein loss combined with gradient penalty. The method has been evaluated on 14 UCI benchmark datasets and achieves state-of-the-art performance in anomaly detection compared to existing advanced algorithms, with an average improvement of 8.3% in AUC, an 11.2% increase in F1-Score, and a 15.7% reduction in false positive rate. Full article