Search Results (12,000)

As global agriculture shifts to intelligence and precision, crop attribute detection has become foundational for intelligent systems (harvesters, UAVs, sorters). It enables real-time monitoring of key indicators (maturity, moisture, disease) to optimize operations—reducing crop losses by 10–15% via precise cutting height adjustment—and boosts resource-use efficiency. This review targets harvesting-stage and in-field monitoring for grains, fruits, and vegetables, highlighting practical technologies: near-infrared/Raman spectroscopy (non-destructive internal attribute detection), 3D vision/LiDAR (high-precision plant height/density/fruit location measurement), and deep learning (YOLO for counting, U-Net for disease segmentation). It addresses universal field challenges (lighting variation, target occlusion, real-time demands) and actionable fixes (illumination compensation, sensor fusion, lightweight AI) to enhance stability across scenarios. Future trends prioritize real-world deployment: multi-sensor fusion (e.g., RGB + thermal imaging) for comprehensive perception, edge computing (inference delay < 100 ms) to solve rural network latency, and low-cost solutions (mobile/embedded device compatibility) to lower smallholder barriers—directly supporting scalable precision agriculture and global sustainable food production. Full article

In autonomous driving, detecting small and occluded objects remains a substantial challenge due to the complexity of real-world environments. To address this, we propose RSO-YOLO, an enhanced model based on YOLOv12. First, the bidirectional feature pyramid network (BiFPN) and space-to-depth convolution (SPD-Conv) replace the original neck network. This design efficiently integrates multi-scale features while preserving fine-grained information during downsampling, thereby improving both computational efficiency and detection performance. Additionally, a detection head for the shallower feature layer P2 is incorporated, further boosting the model’s capability to detect small objects. Second, we propose the feature enhancement and compensation module (FECM), which strengthens features in visible regions and compensates for missing semantic information in occluded areas. This module improves detection accuracy and robustness under occlusion. Finally, we propose a lightweight global cross-dimensional coordinate detection head (GCCHead), built upon the global cross-dimensional coordinate module (GCCM). By grouping and synergistically enhancing features, this module addresses the challenge of balancing computational efficiency with detection performance. Experimental results demonstrate that on the SODA10M, BDD100K, and FLIR ADAS datasets, RSO-YOLO achieves mAP@0.5 improvements of 8.0%, 10.7%, and 7.2%, respectively, compared to YOLOv12. Meanwhile, the number of parameters is reduced by 15.4%, and model complexity decreases by 20%. In summary, RSO-YOLO attains higher detection accuracy while reducing parameters and computational complexity, highlighting its strong potential for practical autonomous driving applications. Full article

Unstable technological processes, such as turbulent gas and hydrodynamic flows, generate time series that deviate sharply from the assumptions of classical statistical forecasting. These signals are shaped by stochastic chaos, characterized by weak inertia, abrupt trend reversals, and pronounced low-frequency contamination. Traditional extrapolators, including linear and polynomial models, therefore act only as weak forecasters, introducing systematic phase lag and rapidly losing directional reliability. To address these challenges, this study introduces an evolutionary boosting framework within a multi-expert system (MES) architecture. Each expert is defined by a compact genome encoding training-window length and polynomial order, and experts evolve across generations through variation, mutation, and selection. Unlike conventional boosting, which adapts only weights, evolutionary boosting adapts both the weights and the structure of the expert pool, allowing the system to escape local optima and remain responsive to rapid environmental shifts. Numerical experiments on real monitoring data demonstrate consistent error reduction, highlighting the advantage of short windows and moderate polynomial orders in balancing responsiveness with robustness. The results show that evolutionary boosting transforms weak extrapolators into a strong short-horizon forecaster, offering a lightweight and interpretable tool for proactive control in environments dominated by chaotic dynamics. Full article

Researching cosmic dust requires terrestrial facilities for accelerating analogues of different sizes and masses. To address the area of very lightweight particles, electrostatic accelerators like Van de Graaf accelerators or Linear Accelerators (LINACs) have proven adequate. This article describes the components, dimensions, working principle and attributes of a variable frequency switched 6-stage LINAC of 120 kilovolts (kV) potential based at the Institute of Space Systems, University of Stuttgart. It utilizes negative voltages, no storage capacitors, isometric drift tubes, one semiconductor-based high-voltage switch per stage and there is no voltage drop during acceleration. The particle rate can reach up to 33 particles per second. By setting a target speed window, it autonomously chooses the right number of acceleration stages to meet that requirement, if possible. Micron-sized iron particles were accelerated successfully, achieving speed increase rates of up to three times the pre-LINAC speed and a total speed of up to 1300 m per second (m/s). This platform provides a new tool for dust sensor calibration, impact physics and material surface processing due to its ability to bring particles of different charge-to-mass ratios to a defined target speed. Full article

The European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis v5 (ERA5) datasets provide a rich source of climatological data. However, their Network Common Data Form (NetCDF) structure can be a barrier for researchers who are not experienced with specialized data tools or programming languages. To address this challenge, we developed ncPick, a lightweight, Windows-based application designed to make ERA5 data more accessible and easier to use. The software enables users to load NetCDF files, select points of interest manually or through shapefiles, and export the data directly to Comma-separated values (CSV) format for further processing in common tools such as Excel, R, or within ncPick itself. Additional modules allow for quick visualization, descriptive statistics, interpolation, and the generation of time-of-day heatmaps, as well as practical data handling functions such as merging and downsampling CSV files based on the time-axis. Validation tests confirmed that ncPick outputs are consistent with those from established tools (such as Panoply). The toolkit was found to be stable across different Windows systems and suitable for a range of datasets. While it has limitations with very large files and does not include automated data download for version 1 of the software, ncPick offers an accessible solution for researchers, students, and other professionals seeking a reliable and intuitive way to work with ERA5 NetCDF data. Full article

This study deals with the fatigue behavior of on-site-installation-type track slabs subject to cycling train load developed by applying glass-fiber-reinforced polymer (GFRP) reinforcing bars. Concrete track slabs have the most severe deterioration in track circuit characteristic values due to the conduction influence of existing steel bars. Therefore, a track slab applying an insulator and lightweight GFRP reinforcement by replacing the existing steel bar was proposed from a design perspective. In order to present the validity of the proposed method, a full-size specimen was manufactured and a fatigue performance test was performed, and the results were compared with the test specimen applied with steel bars. From the results of various fatigue behaviors, it was found that displacement variations during cyclic loading remained within 1 mm, and load variations were within 10 kN, indicating excellent stability under accumulated fatigue cycles. This study analyzed the macro-level structural behavior of GFRP-reinforced concrete track slabs under fatigue loading. Future research will further investigate micro-level bond interactions between the reinforcement and concrete to validate long-term performance. Full article

Large Area Additive Manufacturing (LAAM) enables the rapid production of thermoplastic polymer structures but suffers from significant anisotropy and 3D printability limitations. These limitations often require additional material and time in order to incorporate supporting structures. This research explores the integration of continuous fiber reinforced thermoplastics (CFRTP) with LAAM structures. A series of experimental trials were performed, which demonstrate the feasibility and benefits of CFRTP integration, as it can improve structural strength, lightweighting, and manufacturing flexibility. The findings suggest that CFRTP integration can significantly enhance LAAM by reducing material usage, improving mechanical properties, and expanding design possibilities. While further research is needed to optimize the process for specific applications, this process of Hybrid Advanced Additive Manufacturing (HAAM) presents a promising approach for advancing large-scale additive manufacturing. Full article

This study presents an analytical investigation of the thermomechanical stability of hyperbolic doubly curved shells reinforced with graphene origami auxetic metamaterials (GOAMs) and resting on a Pasternak elastic foundation. The proposed model integrates shell geometry, thermal–mechanical loading, and architected auxetic reinforcement to capture their coupled influence on buckling behavior. Stability equations are derived using the First-Order Shear Deformation Theory (FSDT) and the principle of virtual work, while the effective thermoelastic properties of the GOAM phase are obtained through micromechanical homogenization as functions of folding angle, mass fraction, and spatial distribution. Closed-form eigenvalue solutions are achieved with Navier’s method for simply supported boundaries. The results reveal that GOAM reinforcement enhances the critical buckling load at low folding angles, whereas higher folding induces compliance that diminishes stability. The Pasternak shear layer significantly improves buckling resistance up to about 46% with pronounced effects in asymmetrically graded configurations. Compared with conventional composite shells, the proposed GOAM-reinforced shells exhibit tunable, folding-dependent stability responses. These findings highlight the potential of origami-inspired graphene metamaterials for designing lightweight, thermally stable thin-walled structures in aerospace morphing skins and multifunctional mechanical systems. Full article

This study presents a novel federated learning (FL) methodology implemented directly on STM32-based microcontrollers (MCUs) for energy-efficient smart irrigation. To the best of our knowledge, this is the first work to demonstrate end-to-end FL training and aggregation on real STM32 MCU clients (STM32F722ZE), under realistic energy and memory constraints. Unlike most prior studies that rely on simulated clients or high-power edge devices, our framework deploys lightweight neural networks trained locally on MCUs and synchronized via message queuing telemetry transport (MQTT) communication. Using a smart agriculture (SA) dataset partitioned by soil type, 7 clients collaboratively trained a model over 3 federated rounds. Experimental results show that MCU clients achieved competitive accuracy (70–82%) compared to PC clients (80–85%) while consuming orders of magnitude less energy. Specifically, MCU inference required only 0.95 mJ per sample versus 60–70 mJ on PCs, and training consumed ∼70 mJ per epoch versus nearly 20 J. Latency remained modest, with MCU inference averaging 3.2 ms per sample compared to sub-millisecond execution on PCs, a negligible overhead in irrigation scenarios. The evaluation also considers the payoff between accuracy, energy consumption, and latency through the Energy Latency Accuracy Index (ELAI). This integrated perspective highlights the trade-offs inherent in deploying FL on heterogeneous devices and demonstrates the efficiency advantages of MCU-based training in energy-constrained smart irrigation settings. Full article

Acquiring high-resolution digital elevation models (DEMs) over across extensive regions remains challenging due to high costs and insufficient detail, creating demand for super-resolution (SR) techniques. However, existing DEM SR methods still rely on limited data sources and often neglect essential terrain features. To address the issues, SAR data complements existing sources with its all-weather capability and strong penetration, and a Transformer-based Residual Attention Network combining SAR and Terrain Features (TRAN-ST) is proposed. The network incorporates intensity and coherence as SAR features to restore the details of the high-resolution DEMs, while slope and aspect constraints in the loss function enhance terrain consistency. Additionally, it combines the lightweight Transformer module with the residual feature aggregation module, which enhances the global perception capability while aggregating local residual features, thereby improving the reconstruction accuracy and training efficiency. Experiments were conducted on two DEMs in San Diego, USA, and the results show that compared with methods such as the bicubic, SRCNN, EDSR, RFAN, HNCT methods, the model reduces the mean absolute error (MAE) by 2–30%, the root mean square error (RMSE) by 1–31%, and the MAE of the slope by 2–13%, and it reduces the number of parameters effectively, which proves that TRAN-ST outperforms current typical methods. Full article

Active exosuits provide various assistive force profiles but are limited by battery life, weight, and complex maintenance requirements. Passive exosuits, by contrast, are economical and lightweight while also offering unlimited usage times; however, due to their fixed stiffness levels, they can provide only a limited set of optimized assistive force profiles for different movements. To address these issues, this paper proposes a new variable elastic band for semi-passive exosuits. It comprises rubber bands and webbings connected in parallel, with the elongation of the rubber bands restricted according to the webbing length. By connecting these segments in series, a range of elasticities can be generated. Experimental results confirmed that the band could generate different stiffness levels, which were accurately predicted with an average coefficient of determination (R²) of 0.9985 and an average root mean square error of 0.8993. Additionally, based on tests involving participants wearing the device, the variable elastic band effectively modulated the assistive force profile. These findings overcome the previous limitations of passive components, opening the door to future research on enhancing the efficiency of passive systems and enabling further customization. Full article

This paper presents the design and testing of a lightweight, low-cost robotic arm with an extended vertical range. The 9-degree-of-freedom (DOF) system comprises a 6-DOF arm and a 3-DOF gripper. To minimize weight, the six wrist and gripper joints are cable-driven, with all actuators relocated to the shoulder assembly. As a result, the wrist and gripper only weigh 222 g and 113 g, respectively, significantly reducing the inertia on the end effector. The arm utilizes a SCARA-configuration that slides along a tower for extended vertical reach. A key innovation is a closed-section frame that attaches the arm to the tower, in which the bending and torsional loads from the payload can be directly transferred onto the static structure. In contrast to conventional design, this design does not require the shoulder motor to take the bending load directly. Instead, the motor only needs to overcome the rolling friction of the reaction load. Experimental results demonstrate that this approach reduces the required motor torque by a factor of 30. Consequently, the prototype can manipulate a 3 kg payload at a 0.5 m lateral reach while weighing only 4.5 kg, costing USD 1200, and consuming a maximum of 11.1 W of power. Full article

Background: Renal Cell Carcinoma (RCC) is the most common type of kidney cancer and requires accurate histopathological grading for effective prognosis and treatment planning. However, manual grading is time-consuming, subjective, and susceptible to inter-observer variability. Objective: This study proposes EAT-Net (Efficient Attention Transformer Network), a dual-stream deep learning model designed to automate and enhance RCC grade classification from histopathological images. Method: EAT-Net integrates EfficientNetB0 for local feature extraction and a Vision Transformer (ViT) stream for capturing global contextual dependencies. The architecture incorporates Squeeze-and-Excitation (SE) modules to recalibrate feature maps, improving focus on informative regions. The model was trained and evaluated on two publicly available datasets, KMC-RENAL and RCCG-Net. Standard preprocessing was applied, and the model’s performance was assessed using accuracy, precision, recall, and F1-score. Results: EAT-Net achieved superior results compared to state-of-the-art models, with an accuracy of 92.25%, precision of 92.15%, recall of 92.12%, and F1-score of 92.25%. Ablation studies demonstrated the complementary value of the EfficientNet and ViT streams. Additionally, Grad-CAM visualizations confirmed that the model focuses on diagnostically relevant areas, supporting its interpretability and clinical relevance. Conclusion: EAT-Net offers an accurate, and explainable framework for RCC grading. Its lightweight architecture and high performance make it well-suited for clinical deployment in digital pathology workflows. Full article

Lightweight Single-Image Super-Resolution (SISR) faces the core challenge of balancing computational efficiency with reconstruction quality, particularly in preserving both high-frequency details and global structures under constrained resources. To address this, we propose the Periodically Enhanced Cascade Network (PECNet). Our main contributions are as follows: 1. Its core component, a novel Multi-scale Adaptive Feature Aggregation (MAFA) module, which employs three functionally complementary branches that work synergistically: one dedicated to extracting local high-frequency details, another to efficiently modeling long-range dependencies and a third to capturing structured contextual information within windows. 2. To seamlessly integrate these branches and enable cross-window information interaction, we introduce the Periodic Boundary Padding Shift (PBPS) mechanism. This mechanism serves as a symmetric preprocessing step that achieves implicit window shifting without introducing any additional computational overhead. Extensive benchmarking shows PECNet achieves better reconstruction quality without a complexity increase. Taking the representative shift-window-based lightweight model, NGswin, as an example, for ×4 SR on the Manga109 dataset, PECNet achieves an average PSNR 0.25 dB higher, while its computational cost (in FLOPs) constitutes merely 40% of NGswin’s. Full article

Image enhancement in complex scenes is challenging due to the frequent coexistence of multiple degradations caused by adverse weather, imaging hardware, and transmission environments. Existing datasets remain limited to single or weather-specific degradation types, failing to capture real-world complexity. To address this gap, we introduce the Robust Multi-Type Degradation (RMTD) dataset, which synthesizes a wide range of degradations from meteorological, capture, and transmission sources to support model training and evaluation under realistic conditions. Furthermore, the superposition of multiple degradations often results in feature maps dominated by noise, obscuring underlying clean content. To tackle this, we propose the Feature Filter Transformer (FFformer), which includes: (1) a Gaussian-Filtered Self-Attention (GFSA) module that suppresses degradation-related activations by integrating Gaussian filtering into self-attention; and (2) a Feature-Shrinkage Feed-forward Network (FSFN) that applies soft-thresholding to aggressively reduce noise. Additionally, a Feature Enhancement Block (FEB) embedded in skip connections further reinforces clean background features to ensure high-fidelity restoration. Extensive experiments on RMTD and public benchmarks confirm that the proposed dataset and FFformer together bring substantial improvements to the task of complex-scene image enhancement. Full article