Search Results (29,515)

Soluble solids content is the most important attribute related to the quality and price of apples. The objective of this study was to detect the soluble solids content (SSC) in ‘Fuji’ apples using hyperspectral imaging combined with a deep learning algorithm. The hyperspectral images of 570 apple samples were obtained and the whole region of apple sample hyperspectral data was collected and preprocessed. In addition, a method involving multi-attention convolutional neural network (MA-CNN) is proposed, which extracts spectral and spatial features from hyperspectral images by embedding channel attention (CA) and spatial attention (SA) modules in a convolutional neural network. The CA and SA modules help the network adaptively focus on important spectral–spatial features while reducing the interference of redundant information. Additionally, the Bayesian optimization algorithm (BOA) is used for model hyperparameter optimization. A comprehensive evaluation is conducted by comparing the proposed model with CA-CNN models, SA-CNN, and the current mainstream models. Furthermore, the best prediction performances for detecting SSC in apple samples were obtained from the MA-CNN model, with an ${\mathrm{R}}_{\mathrm{p}}^2$ value of 0.9602 and an RMSEP value of 0.0612 °Brix. The results of this study indicated that the MA-CNN algorithm combined with hyperspectral imaging technology can be used as an effective method for rapid detection of apple quality parameters. Full article

Hyperspectral image (HSI) classification requires models that can simultaneously capture spatial structures and spectral continuity. Although state space models (SSMs), particularly Mamba, have shown strong capability in long-sequence modeling, their application to HSI remains limited due to insufficient spectral relation modeling and the constraints of unidirectional processing. To address these challenges, we propose BiMambaHSI, a novel bidirectional spectral-–spatial framework. First, we proposed a joint spectral–-spatial gated mamba (JGM) encoder that applies forward–backward state modeling with input-dependent gating, explicitly capturing bidirectional spectral–-spatial dependencies. This bidirectional mechanism explicitly captures long-range spectral–-spatial dependencies, overcoming the limitations of conventional unidirectional Mamba. Second, we introduced the spatial-–spectral mamba block (SSMB), which employs parallel bidirectional branches to extract spatial and spectral features separately and integrates them through a lightweight adaptive fusion mechanism. This design enhanced spectral continuity, spatial discrimination, and cross-dimensional interactions while preserving the linear complexity of pure SSMs. Extensive experiments on five public benchmark datasets (Pavia University, Houston, Indian Pines, WHU-Hi-HanChuan, and WHU-Hi-LongKou) demonstrate that BiMambaHSI consistently achieves state-of-the-art performance, improving classification accuracy and robustness compared with existing CNN- and Transformer-based methods. Full article

Referring Remote Sensing Image Segmentation (RRSIS) aims to accurately locate and segment target objects in high-resolution aerial imagery based on natural language descriptions. Most existing approaches either directly modify Referring Image Segmentation (RIS) frameworks originally designed for natural images or employ image-based foundation models such as SAM to improve segmentation accuracy. However, current RRSIS models still face substantial challenges due to the domain gap between remote sensing and natural images, including large-scale variations, arbitrary object rotations, and complex spatial–linguistic relationships. Consequently, such transfers often lead to weak cross-modal interaction, inaccurate semantic alignment, and reduced localization precision, particularly for small or rotated objects. In addition, approaches that rely on multi-stage alignment pipelines, redundant high-level feature fusion, or the incorporation of large foundation models generally incur substantial computational overhead and training inefficiency, especially when dealing with complex referring expressions in high-resolution remote sensing imagery. To address these challenges, we propose CD2FSAN, a CLIP-driven dynamic feature selection and alignment network that establishes a unified framework for fine-grained cross-modal understanding in remote sensing imagery. This network first follows the principle of maximizing cross-modal information to dynamically select the visual representations most semantically aligned with the language from CLIP’s hierarchical features, thereby strengthening cross-modal correspondence under image domain shifts. It then performs adaptive multi-scale aggregation and alignment to integrate linguistic cues into spatially diverse visual contexts, enabling precise feature fusion across varying object scales. Finally, a dynamic rotation correction decoder with differentiable affine transformation was designed to refine segmentation by compensating for orientation diversity and geometric distortions. Extensive experiments verify that CD2FSAN consistently outperforms existing methods in segmentation accuracy, validating the effectiveness of its core components while maintaining competitive computational efficiency. These results demonstrate the framework’s strong capability to bridge the cross-modal gap between language and remote sensing imagery, highlighting its potential for advancing semantic understanding in vision–language remote sensing tasks. Full article

The conventional OLSR protocol faces substantial challenges in highly dynamic and interference-intensive UAV environments, including high mobility, frequent topology changes, and insufficient adaptability to electromagnetic interference. This paper proposes a cross-layer improved OLSR protocol, OLSR-LCN, that integrates three evaluation metrics—link lifetime (LL), channel interference index (CII), and node load (NL)—to enhance communication stability and network performance. The proposed protocol extends the OLSR control message structure and employs enhanced MPR selection and routing path computation algorithms. LL prediction enables proactive selection of stable communication paths, while the CII helps avoid heavily interfered nodes during MPR selection. Additionally, the NL metric facilitates load balancing and prevents premature node failure due to resource exhaustion. Simulation results demonstrate that across different UAV flight speeds and network scales, OLSR-LCN protocol consistently outperforms both the OLSR and the position-based OLSR in terms of end-to-end delay, packet loss rate, and network efficiency. The cross-layer optimization approach effectively addresses frequent link disruptions, interference, and load imbalance in dynamic environments, providing a robust solution for reliable communication in complex FANETs. Full article

Despite the superior performance of existing anomaly detection methods, they are often limited to single-class detection tasks, requiring separate models for each class. This constraint hinders their detection performance and deployment efficiency when applied to real-world multi-class data. In this paper, we propose a dynamic visual adaptation framework for multi-class anomaly detection, enabling the dynamic and adaptive capture of features based on multi-class data, thereby enhancing detection performance. Specifically, our method introduces a network plug-in, the Hyper AD Plug-in, which dynamically adjusts model parameters according to the input data to extract dynamic features. By leveraging the collaboration between the Mamba block, the CNN block, and the proposed Hyper AD Plug-in, we extract global, local, and dynamic features simultaneously. Furthermore, we incorporate the Mixture-of-Experts (MoE) module, which achieves a dynamic balance across different features through its dynamic routing mechanism and multi-expert collaboration. As a result, the proposed method achieves leading accuracy on the MVTec AD and VisA datasets, with image-level mAU-ROC scores of 98.8% and 95.1%, respectively. Full article

Background: The binding of glycosaminoglycans (GAG) to Wnt signaling components plays a key regulatory role in bone formation and regeneration. We previously reported de novo designed chemically modified hyaluronan derivatives, named _REGAG (Rationally Engineered GAG), which demonstrated bone-regenerative properties in a mouse calvaria defect model. To gain initial insights into the pharmacological profile of two _REGAG currently under preclinical investigation in mice, we performed a comprehensive in silico investigation of their binding to human and murine serum albumin (HSA and MSA), as it might influence their ADME properties. Furthermore, we evaluated whether _REGAG binding might impact the recognition of well-characterized HSA-binding drugs. Methods: State-of-the-art in silico ADMET tools, docking and molecular dynamics simulations were used to predict and characterize the interaction of _REGAG with HSA and MSA, and to investigate the molecular mechanisms involved at the atomic level. Results: The investigated _REGAG molecules show a consistent binding preference for the FA1 site in both proteins, and an additional preference for the FA7 site in HSA. Their recognition might induce protein conformational changes and alter the functional state. Furthermore, _REGAG’s conformational adaptability is predicted to influence their binding to the FA5/6 and FA8/9 sites of HSA, and to the FA3/4 and FA7 sites of MSA. Conclusions: Our investigations predict the binding of two hyaluronan derivatives to HSA and MSA. The mechanistic insights gained into the molecular recognition of these two _REGAG molecules offer valuable information for their potential clinical application and serve as a rational basis for future molecular design aimed at improving pharmacokinetic properties. Full article

Urban traffic congestion leads to daily delays, driven by outdated, rigid control systems. As vehicle numbers grow, fixed-phase signals struggle to adapt to real-time conditions. This work presents a decentralized Multi-Agent Reinforcement Learning (MARL) system to manage a traffic cell composed of five intersections, introducing the novel Strategic Anti-Blocking Phase Adjustment (SAPA) module, developed to enable dynamic phase time adjustments. The goal is to optimize arterial traffic flow by adapting strategies to different traffic generation patterns, simulating priority movements along circular or radial arterials, such as inbound or outbound city flows. The system aims to manage diverse scenarios within a cell, with the long-term goal of scaling to city-wide networks. A Visible Light Communication (VLC) infrastructure is integrated to support real-time data exchange between vehicles and infrastructure, capturing vehicle position, speed, and pedestrian presence at intersections. The system is evaluated through multiple performance metrics, showing promising results: reduced vehicle queues and waiting times, increased average speeds, and improved pedestrian safety and overall flow management. These outcomes demonstrate the system’s potential to deliver adaptive, intelligent traffic control for complex urban environments. Full article

To address the challenges in preventing and controlling mastitis caused by Escherichia coli and Staphylococcus aureus in large-scale dairy farms, as well as the issues of traditional disinfection protocols relying on experience and exhibiting significant efficacy fluctuations, this study aimed to systematically explore optimal disinfection strategies adapted to different scenarios and seasons. Five common commercial disinfectants in China were selected to target the two aforementioned pathogenic strains. Experiments were conducted under three typical scenarios—bacterial suspension, stainless steel carriers (simulating milking equipment), and cow dung cubicle bedding—and three temperature conditions (4 °C, 25 °C, 37 °C, simulating seasonal temperatures). A series of tests were performed, including neutralizer identification tests, determination of minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC), quantitative suspension and carrier spray disinfection tests, and monitoring of bacterial growth and decline in cow dung cubicle bedding. These tests were used to quantitatively analyze the regulatory mechanisms of disinfectant concentration, action time, and environmental temperature on disinfection efficacy. The Compound Glutaral Solution (CGS) exhibited the best overall performance, with strong temperature stability across all scenarios and high-efficiency bactericidal activity even at low concentrations. Additionally, the combined system of the CGS and bleaching powder (BP) achieved the optimal effect in controlling bacterial rebound in the cow dung cubicle bedding scenario. This study clarified the scenario-specific adaptation rules of different disinfectants and established a scenario-specific precision disinfection strategy for dairy farms. It provides a scientific basis for improving the level of mastitis prevention and control and optimizing biosafety systems, while also offering references for the disinfection of hard surfaces in fields such as healthcare and food processing. Full article

The separable heat pipe, with its highly efficient heat transfer and flexible layout features, has become an innovative solution to the heat dissipation problem of batteries, especially suitable for the directional heat dissipation requirements of high-energy-density battery packs. However, most of the number–value models currently studied examine the flow of refrigerant working medium within the pump as an isentropic or isothermal process and are unable to effectively analyze the heat transfer characteristics of different internal regions. Based on the laws of energy conservation, momentum conservation, and mass conservation, this study establishes a steady-state mathematical model of the pump-driven microchannel-separated heat pipe. The influence of factors—such as the phase state change in the working medium inside the heat exchanger, the heat transfer flow mechanism, the liquid filling rate, the temperature difference, as well as the structural parameters of the microchannel heat exchanger on the steady-state heat transfer and flow performance of the pump-driven microchannel-separated heat pipe—were analyzed. It was found that the influence of liquid filling ratio on heat transfer quantity is reflected in the ratio of change in the sensible heat transfer and latent heat transfer. The sensible heat transfer ratio is higher when the liquid filling is too low or too high, and the two-phase heat transfer is higher when the liquid filling ratio is in the optimal range; the maximum heat transfer quantity can reach 3.79 KW. The decrease in heat transfer coefficient with tube length in the single-phase region is due to temperature and inlet effect, and the decrease in heat transfer coefficient in the two-phase region is due to the change in flow pattern and heat transfer mechanism. This technology has the advantages of long-distance heat transfer, which can adapt to the distributed heat dissipation needs of large-energy-storage power plants and help reduce the overall lifecycle cost. Full article

This study proposes an adaptive fault-tolerant control strategy for parametric strict-feedback systems subject to actuator faults and unknown dead-zone nonlinearities, a combination that presents significant challenges for controller design. First, a novel prescribed-performance fault-tolerant control framework is developed by incorporating a funnel function, a barrier Lyapunov function, and a bounded estimation mechanism to address the issue of multiple constrained nonlinear disturbances. Second, the proposed strategy offers two key improvements: (1) adequate compensation for the coupled effects of actuator faults and dead-zone nonlinearities, and (2) guaranteed globally prescribed transient performance, making the settling time and tracking accuracy independent of initial conditions and design parameters. Lastly, simulation results verify the approach’s effectiveness, showing rapid convergence within 0.8 s and a tracking error bounded by $\pm 0.05$, thus surpassing traditional methods. Full article

In recent years, animal welfare has become a high priority in livestock production systems owing to the pressure to balance environmental sustainability, productivity, and ethics as demand continues to grow. This review presents the latest advances in small ruminant welfare, with emphasis on the effects of climate change, the main new innovative managerial and husbandry methods, and the use of precision livestock farming (PLF) technologies. In the first part, this review will examine how climate change is already re-shaping environmental and physiological conditions for farmed sheep and goats, with rising heat stress and negative impacts on both productive and reproductive performance. Secondly, more recent advances in small ruminant management will be presented, including improved housing systems, nutritional strategies, and behavioral monitoring, aimed at enhancing animal resilience and performance. Finally, particular focus will be given to the use of PLF tools for assessing milk quality and monitoring animal welfare. Evidence suggests that real-time monitoring technologies and sensor systems can accurately capture physiological and production parameters and provide an early sign of stress or health issues. Overall, the findings suggest that an integrated approach, combining climate adaptation strategies, welfare management, and the integration of precision technologies can serve as a key driver toward more ethical, sustainable, and resilient livestock production systems. Full article

Medical image segmentation, driven by the intrinsic fractal characteristics of biological patterns, plays a crucial role in medical image analysis. Recently, universal image segmentation, which aims to build models that generalize robustly to unseen anatomical structures and imaging modalities, has emerged as a promising research direction. To achieve this, previous solutions typically follow the in-context learning (ICL) framework, leveraging segmentation priors from a few labeled in-context references to improve prediction performance on out-of-distribution samples. However, these ICL-based methods often overlook the quality of the in-context set and struggle with capturing intricate anatomical details, thus limiting their segmentation accuracy. To address these issues, we propose VG-SAM, which employs a multi-scale in-context retrieval phase and a visual in-context guided segmentation phase. Specifically, inspired by the hierarchical and self-similar properties in fractal structures, we introduce a multi-level feature similarity strategy to select in-context samples that closely match the query image, thereby ensuring the quality of the in-context samples. In the segmentation phase, we propose to generate multi-granularity visual prompts based on the high-quality priors from the selected in-context set. Following this, these visual prompts, along with the semantic guidance signal derived from the in-context set, are seamlessly integrated into an adaptive fusion module, which effectively guides the Segment Anything Model (SAM) with powerful segmentation capabilities to achieve accurate predictions on out-of-distribution query images. Extensive experiments across multiple datasets demonstrate the effectiveness and superiority of our VG-SAM over the state-of-the-art (SOTA) methods. Notably, under the challenging one-shot reference setting, our VG-SAM surpasses SOTA methods by an average of $6.61\%$ in DSC across all datasets. Full article

CCTV safety monitoring demands anomaly detectors combine reliable clip-level accuracy with predictable per-clip latency despite weak supervision. This work investigates compact vision–language models (VLMs) as practical detectors for this regime. A unified evaluation protocol standardizes preprocessing, prompting, dataset splits, metrics, and runtime settings to compare parameter-efficiently adapted compact VLMs against training-free VLM pipelines and weakly supervised baselines. Evaluation spans accuracy, precision, recall, F1, ROC-AUC, and average per-clip latency to jointly quantify detection quality and efficiency. With parameter-efficient adaptation, compact VLMs achieve performance on par with, and in several cases exceeding, established approaches while retaining competitive per-clip latency. Adaptation further reduces prompt sensitivity, producing more consistent behavior across prompt regimes under the shared protocol. These results show that parameter-efficient fine-tuning enables compact VLMs to serve as dependable clip-level anomaly detectors, yielding a favorable accuracy–efficiency trade-off within a transparent and consistent experimental setup. Full article

In-band full-duplex (IBFD) communication systems offer a promising means of improving spectral efficiency by enabling simultaneous transmission and reception on the same frequency channel. Despite this advantage, self-interference (SI) remains a major challenge to their practical deployment. Among the different SI cancellation (SIC) techniques, this paper focuses on digital SIC methodologies tailored for multiple-input multiple-output (MIMO) wireless transceivers operating under digital beamforming architectures. Two distinct digital SIC approaches are evaluated, employing a generalized memory polynomial (GMP) model augmented with Itô–Hermite polynomial basis functions and a phase-normalized neural network (PNN) to effectively model the nonlinearities and memory effects introduced by transmitter and receiver hardware impairments. The robustness of the SIC is further evaluated under both single off-line training and closed-loop real-time adaptation, employing estimation techniques such as least squares (LS), least mean squares (LMS), and fast Kalman (FK) for model coefficient estimation. The performance of the proposed digital SIC techniques is evaluated through detailed simulations that incorporate realistic power amplifier (PA) characteristics, channel conditions, and high-order modulation schemes. Metrics such as error vector magnitude (EVM) and total bit error rate (BER) are used to assess the quality of the received signal after SIC under different signal-to-interference ratio (SIR) and signal-to-noise ratio (SNR) conditions. The results show that, for time-variant scenarios, a low-complexity adaptive SIC can be realized using a GMP model with FK parameter estimation. However, in time-invariant scenarios, an open-loop SIC approach based on PNN offers superior performance and maintains robustness across various modulation schemes. Full article

To achieve accurate estimates of a lithium-ion battery’s charge level (SOC) and health condition (SOH), this paper tackles key issues in battery management by introducing a framework built around an adaptive square root cubature Kalman filter (ASRCKF) that tracks parameters in real time for better performance in changing environments. It uses ASRCKF to gauge SOC, while an extended Kalman filter (EKF) identifies battery traits online and monitors capacity loss, with a two-way feedback system that feeds SOH updates directly into the SOC calculations. Testing in high-speed driving, the New European Driving Cycle, and urban stop–start conditions showed the method keeps average SOC errors to 0.16% at most and peak errors to 0.33%, beating out standard EKF and SRCKF approaches in accuracy; SOH errors averaged 0.42%. Overall, this setup proves reliable for combined SOC-SOH tracking in diverse real-world situations, helping to ensure safer battery operations. Full article