Search Results (2,821)

Accurate and timely estimation of soybean emergence at the plot scale using unmanned aerial vehicle (UAV) remote sensing imagery is essential for germplasm evaluation in breeding programs, where breeders prioritize overall plot-scale emergence rates over subimage-based counts. This study proposes PlotCounter, a deep learning regression model based on the TasselNetV2++ architecture, designed for plot-scale soybean seedling counting. It employs a patch-based training strategy combined with full-plot validation to achieve reliable performance with limited breeding plot data. To incorporate additional agronomic information, PlotCounter is extended into a multitask learning framework (MTL-PlotCounter) that integrates sowing metadata such as variety, number of seeds per hole, and sowing density as auxiliary classification tasks. RGB images of 54 breeding plots were captured in 2023 using a DJI Mavic 2 Pro UAV and processed into an orthomosaic for model development and evaluation, showing effective performance. PlotCounter achieves a root mean square error (RMSE) of 6.98 and a relative RMSE (rRMSE) of 6.93%. The variety-integrated MTL-PlotCounter, V-MTL-PlotCounter, performs the best, with relative reductions of 8.74% in RMSE and 3.03% in rRMSE compared to PlotCounter, and outperforms representative YOLO-based models. Additionally, both PlotCounter and V-MTL-PlotCounter are deployed on a web-based platform, enabling users to upload images via an interactive interface, automatically count seedlings, and analyze plot-scale emergence, powered by a multimodal large language model. This study highlights the potential of integrating UAV remote sensing, agronomic metadata, specialized deep learning models, and multimodal large language models for advanced crop monitoring. Full article

Residential green space is an important component of urban green space and one of the major indicators for evaluating the quality of a residential community. Traditional indicators such as the green space ratio only consider the relationship between green space area and total area of the residential community while ignoring the difference in the amount of green space enjoyed by household residents in high-rise and low-rise buildings. Therefore, it is meaningful to estimate household green space and its spatial distribution in residential communities. However, there are frequent difficulties in obtaining specific green space area and household number through ground surveys or consulting with property management units. In this study, taking a composite residential community in Chongqing, China, as the study site, we first employed a five-lens drone to capture its oblique RGB images and generated the DOM (Digital Orthophoto Map). Subsequently, the green space area and distribution in the entire residential community were extracted from the DOM using VDVI (Visible Difference Vegetation Index). The YOLACT (You Only Look At Coefficients) instance segmentation model was used to recognize balconies from the facade images of high-rise buildings to determine their household numbers. Finally, the average green space per household in the entire residential community was calculated to be 67.82 m², and those in the high-rise and low-rise building zones were 51.28 m² and 300 m², respectively. Compared with the green space ratios of 65.5% and 50%, household green space more truly reflected the actual green space occupation in high- and low-rise building zones. Full article

Monitoring sediment concentration in water bodies is crucial for assessing water quality, ecosystems, and environmental health. However, physical sampling and sensor-based approaches are labor-intensive and unsuitable for large-scale, continuous monitoring. This study employs machine learning models to estimate suspended sediment concentration using images captured in natural light, named RGB, and near-infrared (NIR) conditions. A controlled dataset of approximately 1300 images with SSC values ranging from 1000 mg/L to 150,000 mg/L was developed, incorporating temperature, time of image capture, and solar irradiance as additional features. Random forest regression and gradient boosting regression were trained on mean RGB values, red reflectance, time of captured, and temperature for natural light images, achieving up to 72.96% accuracy within a 30% relative error. In contrast, NIR images leveraged gray-level co-occurrence matrix texture features and temperature, reaching 83.08% accuracy. Comparative analysis showed that ensemble models outperformed deep learning models like Convolutional Neural Networks and Multi-Layer Perceptrons, which struggled with high-dimensional feature extraction. These findings suggest that using machine learning models and RGB and NIR imagery offers a scalable, non-invasive, and cost-effective way of sediment monitoring in support of water quality assessment and environmental management. Full article

To address the limitations of the existing cotton seedling growth prediction methods in field environments, specifically, poor representation of spatiotemporal features and low visual fidelity in texture rendering, this paper proposes an algorithm for the prediction of cotton seedling growth from images based on FCA-STNet. The model leverages historical sequences of cotton seedling RGB images to generate an image of the predicted growth at time t + 1 and extracts 37 phenotypic traits from the predicted image. A novel STNet structure is designed to enhance the representation of spatiotemporal dependencies, while an Adaptive Fine-Grained Channel Attention (FCA) module is integrated to capture both global and local feature information. This attention mechanism focuses on individual cotton plants and their textural characteristics, effectively reducing the interference from common field-related challenges such as insufficient lighting, leaf fluttering, and wind disturbances. The experimental results demonstrate that the predicted images achieved an MSE of 0.0086, MAE of 0.0321, SSIM of 0.8339, and PSNR of 20.7011 on the test set, representing improvements of 2.27%, 0.31%, 4.73%, and 11.20%, respectively, over the baseline STNet. The method outperforms several mainstream spatiotemporal prediction models. Furthermore, the majority of the predicted phenotypic traits exhibited correlations with actual measurements with coefficients above 0.8, indicating high prediction accuracy. The proposed FCA-STNet model enables visually realistic prediction of cotton seedling growth in open-field conditions, offering a new perspective for research in growth prediction. Full article

To address the inefficiency of traditional concrete crack detection methods and the heavy reliance of supervised learning on extensive labeled data, in this study, an intelligent assessment method of concrete damage based on pseudo-label semi-supervised learning and fractal geometry theory is proposed to solve two core tasks: one is binary classification of pixel-level cracks, and the other is multi-category assessment of damage state based on crack morphology. Using three-channel RGB images as input, a dual-path collaborative training framework based on U-Net encoder–decoder architecture is constructed, and a binary segmentation mask of the same size is output to achieve the accurate segmentation of cracks at the pixel level. By constructing a dual-path collaborative training framework and employing a dynamic pseudo-label refinement mechanism, the model achieves an F1-score of 0.883 using only 50% labeled data—a mere 1.3% decrease compared to the fully supervised benchmark DeepCrack (F1 = 0.896)—while reducing manual annotation costs by over 60%. Furthermore, a quantitative correlation model between crack fractal characteristics and structural damage severity is established by combining a U-Net segmentation network with the differential box-counting algorithm. The experimental results demonstrate that under a cyclic loading of 147.6–221.4 kN, the fractal dimension monotonically increases from 1.073 (moderate damage) to 1.189 (failure), with 100% accuracy in damage state identification, closely aligning with the degradation trend of macroscopic mechanical properties. In complex crack scenarios, the model attains a recall rate (Re = 0.882), surpassing U-Net by 13.9%, with significantly enhanced edge reconstruction precision. Compared with the mainstream models, this method effectively alleviates the problem of data annotation dependence through a semi-supervised strategy while maintaining high accuracy. It provides an efficient structural health monitoring solution for engineering practice, which is of great value to promote the application of intelligent detection technology in infrastructure operation and maintenance. Full article

Work-related musculoskeletal disorders (WMSDs) are a leading concern in industrial ergonomics, often stemming from sustained non-neutral postures and repetitive tasks. This paper presents a vision-based framework for real-time, frame-level ergonomic risk classification using a lightweight Vision Transformer (ViT). The proposed system operates directly on raw RGB images without requiring skeleton reconstruction, joint angle estimation, or image segmentation. A single ViT model simultaneously classifies eight anatomical regions, enabling efficient multi-label posture assessment. Training is supervised using a multimodal dataset acquired from synchronized RGB video and full-body inertial motion capture, with ergonomic risk labels derived from RULA scores computed on joint kinematics. The system is validated on realistic, simulated industrial tasks that include common challenges such as occlusion and posture variability. Experimental results show that the ViT model achieves state-of-the-art performance, with F1-scores exceeding 0.99 and AUC values above 0.996 across all regions. Compared to previous CNN-based system, the proposed model improves classification accuracy and generalizability while reducing complexity and enabling real-time inference on edge devices. These findings demonstrate the model’s potential for unobtrusive, scalable ergonomic risk monitoring in real-world manufacturing environments. Full article

To address the problems of traditional methods that rely on destructive sampling, the poor adaptability of fixed equipment, and the susceptibility of single-view angle measurements to occlusions, a non-destructive and portable device for three-dimensional phenotyping and biomass detection in lettuce was developed. Based on the Structure-from-Motion Multi-View Stereo (SFM-MVS) algorithms, a high-precision three-dimensional point cloud model was reconstructed from multi-view RGB image sequences, and 12 phenotypic parameters, such as plant height, crown width, were accurately extracted. Through regression analyses of plant height, crown width, and crown height, and the R² values were 0.98, 0.99, and 0.99, respectively, the RMSE values were 2.26 mm, 1.74 mm, and 1.69 mm, respectively. On this basis, four biomass prediction models were developed using Adaptive Boosting (AdaBoost), Support Vector Regression (SVR), Gradient Boosting Decision Tree (GBDT), and Random Forest Regression (RFR). The results indicated that the RFR model based on the projected convex hull area, point cloud convex hull surface area, and projected convex hull perimeter performed the best, with an R² of 0.90, an RMSE of 2.63 g, and an RMSEn of 9.53%, indicating that the RFR was able to accurately simulate lettuce biomass. This research achieves three-dimensional reconstruction and accurate biomass prediction of facility lettuce, and provides a portable and lightweight solution for facility crop growth detection. Full article

As Android malware grows increasingly sophisticated, traditional detection methods struggle to keep pace, creating an urgent need for robust, interpretable, and real-time solutions to safeguard mobile ecosystems. This study introduces YoloMal-XAI, a novel deep learning framework that transforms Android application files into RGB image representations by mapping DEX (Dalvik Executable), Manifest.xml, and Resources.arsc files to distinct color channels. Evaluated on the CICMalDroid2020 dataset using YOLO11 pretrained classification models, YoloMal-XAI achieves 99.87% accuracy in binary classification and 99.56% in multi-class classification (Adware, Banking, Riskware, SMS, and Benign). Compared to ResNet-50, GoogLeNet, and MobileNetV2, YOLO11 offers competitive accuracy with at least 7× faster training over 100 epochs. Against YOLOv8, YOLO11 achieves comparable or superior accuracy while reducing training time by up to 3.5×. Cross-corpus validation using Drebin and CICAndMal2017 further confirms the model’s generalization capability on previously unseen malware. An ablation study highlights the value of integrating DEX, Manifest, and Resources components, with the full RGB configuration consistently delivering the best performance. Explainable AI (XAI) techniques—Grad-CAM, Grad-CAM++, Eigen-CAM, and HiRes-CAM—are employed to interpret model decisions, revealing the DEX segment as the most influential component. These results establish YoloMal-XAI as a scalable, efficient, and interpretable framework for Android malware detection, with strong potential for future deployment on resource-constrained mobile devices. Full article

With the rapid proliferation of real-time digital communication, particularly in multimedia applications, securing transmitted image data has become a vital concern. While chaotic systems have shown strong potential for cryptographic use, most existing approaches rely on low-dimensional, integer-order architectures, limiting their complexity and resistance to attacks. Advances in fractional calculus and memristive technologies offer new avenues for enhancing security through more complex and tunable dynamics. However, the practical deployment of high-dimensional fractional-order memristive chaotic systems in hardware remains underexplored. This study addresses this gap by presenting a secure image transmission system implemented on a field-programmable gate array (FPGA) using a universal high-dimensional memristive chaotic topology with arbitrary-order dynamics. The design leverages four- and five-dimensional hyperchaotic oscillators, analyzed through bifurcation diagrams and Lyapunov exponents. To enable efficient hardware realization, the chaotic dynamics are approximated using the explicit fractional-order Runge–Kutta (EFORK) method with the Caputo fractional derivative, implemented in VHDL. Deployed on the Xilinx Artix-7 AC701 platform, synchronized master–slave chaotic generators drive a multi-stage stream cipher. This encryption process supports both RGB and grayscale images. Evaluation shows strong cryptographic properties: correlation of $-6.1081\times {10}^{-5}$, entropy of 7.9991, NPCR of 99.9776%, UACI of 33.4154%, and a key space of $2^{1344}$, confirming high security and robustness. Full article

Monitoring crop damage caused by wildlife remains a significant challenge in agricultural management, particularly in the case of large-scale monocultures such as maize. The given study presents a semi-automated process for detecting wildlife-induced damage using RGB imagery acquired from unmanned aerial vehicles (UAVs). The method is designed for non-specialist users and is fully integrated within the QGIS platform. The proposed approach involves calculating three vegetation indices—Excess Green (ExG), Green Leaf Index (GLI), and Modified Green-Red Vegetation Index (MGRVI)—based on a standardized orthomosaic generated from RGB images collected via UAV. Subsequently, an unsupervised k-means clustering algorithm was applied to divide the field into five vegetation vigor classes. Within each class, 25% of the pixels with the lowest average index values were preliminarily classified as damaged. A dedicated QGIS plugin enables drone data analysts (Drone Data Analysts—DDAs) to adjust index thresholds, based on visual interpretation, interactively. The method was validated on a 50-hectare maize field, where 7 hectares of damage (15% of the area) were identified. The results indicate a high level of agreement between the automated and manual classifications, with an overall accuracy of 81%. The highest concentration of damage occurred in the “moderate” and “low” vigor zones. Final products included vigor classification maps, binary damage masks, and summary reports in HTML and DOCX formats with visualizations and statistical data. The results confirm the effectiveness and scalability of the proposed RGB-based procedure for crop damage assessment. The method offers a repeatable, cost-effective, and field-operable alternative to multispectral or AI-based approaches, making it suitable for integration with precision agriculture practices and wildlife population management. Full article

Yellow rust (Puccinia striiformis) is a common wheat disease that significantly reduces yields, particularly in seasons with cooler temperatures and frequent rainfall. Early detection is essential for effective control, especially in key wheat-producing regions such as Southern Dobrudja, Bulgaria. This study presents a UAV-based approach for detecting yellow rust using only RGB imagery and deep learning for pixel-based classification. The methodology involves data acquisition, preprocessing through histogram equalization, model training, and evaluation. Among the tested models, a UnetClassifier with ResNet34 backbone achieved the highest accuracy and reliability, enabling clear differentiation between healthy and infected wheat zones. Field experiments confirmed the approach’s potential for identifying infection patterns suitable for precision fungicide application. The model also showed signs of detecting early-stage infections, although further validation is needed due to limited ground-truth data. The proposed solution offers a low-cost, accessible tool for small and medium-sized farms, reducing pesticide use while improving disease monitoring. Future work will aim to refine detection accuracy in low-infection areas and extend the model’s application to other cereal diseases. Full article

Recently, deep learning algorithms have been increasingly applied in construction for activity recognition, particularly for excavators, to automate processes and enhance safety and productivity through continuous monitoring of earthmoving activities. These deep learning algorithms analyze construction videos to classify excavator activities for earthmoving purposes. However, previous studies have solely focused on single-source external videos, which limits the activity recognition capabilities of the deep learning algorithm. This paper introduces a novel multi-modal deep learning-based methodology for recognizing excavator activities, utilizing multi-stream input data. It processes point clouds and RGB images using the two-stream long short-term memory convolutional neural network (CNN-LSTM) method to extract spatiotemporal features, enabling the recognition of excavator activities. A comprehensive dataset comprising 495,000 video frames of synchronized RGB and point cloud data was collected across multiple construction sites under varying conditions. The dataset encompasses five key excavator activities: Approach, Digging, Dumping, Idle, and Leveling. To assess the effectiveness of the proposed method, the performance of the two-stream CNN-LSTM architecture is compared with that of single-stream CNN-LSTM models on the same RGB and point cloud datasets, separately. The results demonstrate that the proposed multi-stream approach achieved an accuracy of 94.67%, outperforming existing state-of-the-art single-stream models, which achieved 90.67% accuracy for the RGB-based model and 92.00% for the point cloud-based model. These findings underscore the potential of the proposed activity recognition method, making it highly effective for automatic real-time monitoring of excavator activities, thereby laying the groundwork for future integration into digital twin systems for proactive maintenance and intelligent equipment management. Full article

Corrosion in maritime ballast tanks is a major driver of maintenance costs and operational risks for maritime assets. Inspections are hampered by complex geometries, hazardous conditions, and the limitations of conventional methods, particularly visual assessment, which struggles with subjectivity, accessibility, and early detection, especially under coatings. This paper critically examines these challenges and explores the potential of Light Detection and Ranging (LiDAR) and Hyperspectral Imaging (HSI) to form the basis of improved inspection approaches. We discuss LiDAR’s utility for accurate 3D mapping and providing a spatial framework and HSI’s potential for objective material identification and surface characterization based on spectral signatures along a wavelength range of 400-1000nm (visible and near infrared). Preliminary findings from laboratory tests are presented, demonstrating the basic feasibility of HSI for differentiating surface conditions (corrosion, coatings, bare metal) and relative coating thickness, alongside LiDAR’s capability for detailed geometric capture. Although these results do not represent a deployable system, they highlight how LiDAR and HSI could address key limitations of current practices and suggest promising directions for future research into integrated sensor-based corrosion assessment strategies. Full article

Imaging techniques have revolutionized cultural heritage analysis, particularly for objects that cannot be sampled. This study investigated the utilization of spectral imaging for the identification of pigments in artifacts from the Arxiu Valencià del Disseny, in conjunction with other portable spectroscopy techniques such as XRF, Raman, FT-NIR, and FT-MIR. Four early 1930s watercolors were examined using point-wise elemental and molecular spectroscopic data for pigment classification. Initially, the data cubes obtained with the spectral camera were processed using various methods. The spectral behavior was analyzed pixel-point, and the reflectance curves were qualitatively compared with a set of standards. Subsequently, a computational approach was applied to the data cube to produce RGB, false-color infrared (IRFC), and principal component (PC) images. Algorithms, such as the Vector Angle (VA) mapper, were also employed to map the pigment spectra. Consequently, 19th-century pigments such as Prussian blue, chrome yellow, and alizarin red were distinguished according to their composition, combining the spatial and spectral dimensions of the data. Elemental analysis and infrared spectroscopy supported these findings. In this context, the use of reflectance imaging spectroscopy (RIS), despite its technical limitations, emerged as an essential tool for the documentation and conservation of design heritage. Full article

To improve the identification of drought-affected areas in winter wheat, this paper proposes a lightweight network called MF-FusionNet based on multimodal fusion of RGB images and vegetation indices (NDVI and EVI). A multimodal dataset covering various drought levels in winter wheat was constructed. To enable deep fusion of modalities, a Lightweight Multimodal Fusion Block (LMFB) was designed, and a Dual-Coordinate Attention Feature Extraction module (DCAFE) was introduced to enhance semantic feature representation and improve drought region identification. To address differences in scale and semantics across network layers, a Cross-Stage Feature Fusion Strategy (CFFS) was proposed to integrate multi-level features and enhance overall performance. The effectiveness of each module was validated through ablation experiments. Compared to traditional single-modal methods, MF-FusionNet achieved higher accuracy, recall, and F1-score—improved by 1.35%, 1.43%, and 1.29%, respectively—reaching 96.71%, 96.71%, and 96.64%. A basis for real-time monitoring and precise irrigation management under winter wheat drought stress was provided by this study. Full article