Search Results (3,470)

The species, quantity, and tree diameter at breast height (DBH) are important indicators for assessing species distribution, individual growth status, and overall health in the forest. The existing tree information collection mainly relies on manual labor, which results in low efficiency and high labor intensity. To address these issues, we propose a method for tree species identification and diameter estimation by combining deep learning algorithms with binocular vision. First, an image acquisition platform is designed and integrated with a weeding machine to capture images during weeding operation. Images of seven types of trees are captured to develop a dataset. Second, a tree species identification model is established based on the YOLOv8n network, achieving 98.5% accuracy, 99.0% recall, and 99.2% mAP. Then, an improved YOLOv8n-seg model is proposed. It simplifies the network by introducing VanillaBlock in the backbone. FasterNet with a CCFM structure is added at the neck to enhance the model’s multi-scale expression capability. The mIoU of the improved model is 93.7%. Finally, the improved YOLOv8n-seg model is combined with binocular vision. After obtaining the segmentation mask of the tree, the spatial position of the two measurement points is calculated, allowing for the measurement of tree diameter. Verification experiments show that the average error for tree diameter ranges from 4.40~6.40 mm, and the proposed error compensation method can reduce diameter errors. This study provides a theoretical foundation and technical support for intelligent collection of tree information. Full article

The work aims to leverage computer vision and artificial intelligence technologies to quantify key components in food catering services. Specifically, it focuses on content identification and portion size estimation in a dining hall setting, typical of corporate and educational settings. An RGB camera is employed to capture the tray delivery process in a self-service restaurant, providing test images for content identification algorithm comparison, using standard evaluation metrics. The approach utilizes the YOLO architecture, a widely recognized deep learning model for object detection and computer vision. The model is trained on labeled image data, and its performance is assessed using a precision–recall curve at a confidence threshold of 0.5, achieving a mean Average Precision (mAP) of 0.873, indicating robust overall performance. The weight estimation procedure combines computer vision techniques to measure food volume using both RGB and depth cameras. Subsequently, density models specific to each food type are applied to estimate the detected food weight. The estimation model’s parameters are calibrated through experiments that generate volume-to-weight conversion tables for different food items. Validation of the system was conducted using rice and chicken, yielding error margins of 5.07% and 3.75%, respectively, demonstrating the feasibility and accuracy of the proposed method. Full article

Permafrost degradation under climate warming has profound implications for ecological processes, hydrology, and human activities. Northeast China, characterized by sporadic and isolated patch permafrost near the southern limit of latitudinal permafrost (SLLP), represents one of the most sensitive and complex permafrost regions. This study aims to improve the reliability of permafrost mapping by incorporating parameter uncertainty into simulations. We developed a Monte Carlo–Temperature at the Top of Permafrost (TTOP) (MC–TTOP) framework that integrates an equilibrium model with Monte Carlo sampling to quantify parameter sensitivity and model uncertainty. Using all-sky daily air temperature data and land use and land cover information, we generated probabilistic estimates of mean annual ground temperature (MAGT), permafrost occurrence probability (PZI), and associated uncertainties. Model validation against borehole observations demonstrated improved accuracy compared with global-scale simulations, with a reduced bias and RMSE. Results reveal that permafrost in Northeast China was relatively stable during 2003–2010 but experienced pronounced degradation after 2011, with the total area decreasing to ~2.79 × 10⁵ km² by 2022. Spatial uncertainty was greatest in transitional zones near the southern boundary, where PZI-based delineations tended to overestimate permafrost extent. Regional comparisons further showed that permafrost in Northeast China is more fragmented and uncertain than that on the Tibetan Plateau, owing to complex snow–vegetation–topography interactions and intensive human disturbances. Overall, the MC-TTOP simulations indicate an accelerated permafrost degradation after 2011, with the highest uncertainty concentrated in southern transitional zones near the SLLP, demonstrating that the MC-TTOP framework provides a robust tool for probabilistic permafrost mapping, offering improved reliability for regional-scale assessments and important insights for future risk evaluation under climate change. Full article

Accurate mapping of hurricane-induced damage is essential for guiding rapid disaster response and long-term recovery planning. This study evaluates the Three-Dimensional Multi-Attributes, Multiscale, Multi-Cloud (3DMASC) framework for semantic classification of pre- and post-hurricane Light Detection and Ranging (LiDAR) data, using Mexico Beach, Florida, as a case study following Hurricane Michael. The goal was to assess the framework’s ability to classify stable landscape features and detect damage-specific classes in a highly complex post-disaster environment. Bitemporal topo-bathymetric LiDAR datasets from 2017 (pre-event) and 2018 (post-event) were processed to extract more than 80 geometric, radiometric, and echo-based features at multiple spatial scales. A Random Forest classifier was trained on a 2.37 km² pre-hurricane area (Zone A) and evaluated on an independent 0.95 km² post-hurricane area (Zone B). Pre-hurricane classification achieved an overall accuracy of 0.9711, with stable classes such as ground, water, and buildings achieving precision and recall exceeding 0.95. Post-hurricane classification maintained similar accuracy; however, damage-related classes exhibited lower performance, with debris reaching an F1-score of 0.77, damaged buildings 0.58, and vehicles recording a recall of only 0.13. These results indicate that the workflow is effective for rapid mapping of persistent structures, with additional refinements needed for detailed damage classification. Misclassifications were concentrated along class boundaries and in structurally ambiguous areas, consistent with known LiDAR limitations in disaster contexts. These results demonstrate the robustness and spatial transferability of the 3DMASC–Random Forest approach for disaster mapping. Integrating multispectral data, improving small-object representation, and incorporating automated debris volume estimation could further enhance classification reliability, enabling faster, more informed post-disaster decision-making. By enabling rapid, accurate damage mapping, this approach supports sustainable disaster recovery, resource-efficient debris management, and resilience planning in hurricane-prone regions. Full article

Spatially quantifying the soil carbon sequestration potential (SCSP) is crucial for targeting climate change mitigation strategies like carbon farming. However, static mapping approaches often fail by assuming that the drivers of soil organic carbon (SOC) are stationary. We hypothesized that the hierarchy of SOC controllers is fundamentally non-stationary, shifting from intrinsic stabilization capacity (pedology) in stable ecosystems to extrinsic flux kinetics (climate) in dynamic systems. We tested this by developing a land-use-specific (LULC; Cropland, Forest land, Grassland) ensemble machine learning (ML) framework to quantify the soil carbon saturation deficit (SCSD) across Croatia’s pedologically diverse landscape on 622 soil samples. The LULC-stratified ensemble models (SVM, RF, CUB) achieved moderate to good predictive accuracy under cross-validation (R² = 0.41–0.60). Crucially, the feature importance analysis (permutation MSE loss) proved our hypothesis: in Forest land, SOC was superiorly controlled by intrinsic capacity (Soil CEC, Soil pH), defining the mineralogical C-saturation “ceiling”; in Grasslands, control shifted to extrinsic C-input kinetics (Precipitation: Bio19, Bio12), which “fuel” the microbial carbon pump (MCP) via root exudation; and in Croplands, the model revealed a hybrid control, limited by remaining intrinsic capacity (CEC, Clay) but strongly influenced by C-loss kinetics (Temperature: Bio08), which regulates microbial carbon use efficiency (CUE). This study demonstrates that LULC-specific dynamic modeling is a prerequisite for accurately mapping SCSP. By identifying soils with both high intrinsic capacity (high CEC/Clay) and high degradation (high SCSD), our data-driven assessment provides a critical tool for spatially targeting carbon farming interventions for maximum climate mitigation return on investment (ROI). Full article

Accurate stand count and growth stage detection are essential for crop monitoring, since traditional methods often overlook field variability, leading to poor management decisions. This study evaluated the performance of the YOLOv9-small model for detecting and counting corn plants under real field conditions. The model was tested across three soil background types, two flight heights (30 and 70 m), and four corn growth stages (V2, V3, V5, and V6). Unmanned aerial vehicle (UAV) imagery was collected from three distinct fields and cropped into 640 × 640 pixels. Datasets were split into training (70%), validation (20%), and testing (10%) datasets. Model performance was assessed using precision, recall, classification loss, and mean average precision of 50% and 50–90%. The results showed that the V3 and V5 stages yielded the highest detection accuracy, with mAP50 values exceeding 85% in conventional tillage fields and slightly lower performance in gray/red-brown conditions due to background interference. Increasing flight height to 70 m reduced accuracy by 8–12%, though precision remained high, particularly at V5, and performance was poorest for V2 and V6. In conclusion, YOLOv9-small is effective for early-stage corn detection, particularly at V3 and V5, with 30 m providing optimal results. However, 70 m may be acceptable at V5 to optimize mapping time. Full article

Uniform application of fertilizers and pesticides continues to dominate global agriculture despite significant spatial variability in soil and crop conditions. This mismatch results in avoidable yield gaps, excessive chemical waste, and environmental pressures, including nutrient leaching and greenhouse gas emissions. The integration of Artificial Intelligence (AI) and Remote Sensing (RS) has emerged as a transformative framework for diagnosing this variability and enabling site-specific, climate-responsive management. This systematic synthesis reviews evidence from 2000–2025 to assess how AI–RS technologies optimize agrochemical efficiency. A comprehensive search across Scopus, Web of Science, IEEE Xplore, ScienceDirect, and Google Scholar were used. Following rigorous screening and quality assessment, 142 studies were selected for detailed analysis. Data extraction focused on sensor platforms (Landsat-8/9, Sentinel-1/2, UAVs), AI approaches (Random Forests, CNNs, Physics-Informed Neural Networks), and operational outcomes. The synthesized data demonstrate that AI–RS systems can predict critical soil attributes, specifically salinity, moisture, and nutrient levels, with 80–97% accuracy in some cases, depending on spectral resolution and algorithm choice. Operational implementations of Variable-Rate Application (VRA) guided by these predictive maps resulted in fertilizer reductions of 15–30%, pesticide use reductions of 20–40%, and improvements in water-use efficiency of 25–40%. In fields with high soil heterogeneity, these precision strategies delivered yield gains of 8–15%. AI–RS technologies have matured from experimental methods into robust tools capable of shifting agrochemical science from reactive, uniform practices to predictive, precise strategies. However, widespread adoption is currently limited by challenges in data standardization, model transferability, and regulatory alignment. Future progress requires the development of interoperable data infrastructures, digital soil twins, and multi-sensor fusion pipelines to position these technologies as central pillars of sustainable agricultural intensification. Full article

This study proposes a unified deep-learning framework for the concurrent classification of Chronic Obstructive Pulmonary Disease (COPD) severity and regression of the FEV1/FVC ratio from chest X-ray (CXR) images. We integrated a ConvNeXt-Large backbone with a Slot Attention mechanism to effectively disentangle and refine disease-relevant features for multi-task learning. Evaluation on a clinical dataset demonstrated that the proposed model with a 5-slot configuration achieved superior performance compared to standard CNN and Vision Transformer baselines. On the independent test set, the model attained an Accuracy of 0.9107, Sensitivity of 0.8603, and Specificity of 0.9324 for three-class severity stratification. Simultaneously, it achieved a Mean Absolute Error (MAE) of 8.2649 and a Mean Squared Error (MSE) of 151.4704, and an ${\mathrm{R}}^2$ of 0.7591 for FEV1/FVC ratio estimation. Qualitative analysis using saliency maps also suggested that the slot-based approach contributes to attention patterns that are more constrained to clinically relevant pulmonary structures. These results suggest that our slot-attention-based multi-task model offers a robust solution for automated COPD assessment from standard radiographs. Full article

Lentigo maligna (LM) is a melanoma in situ with high cumulative sun damage. Histological evaluation of resection margins is difficult and time-consuming. Melanocyte density (MD) is a suitable, quantifiable, and reproducible diagnostic criterion. In this retrospective single-centre study, we investigated whether an artificial intelligence (AI) tool can support the assessment of LM. Training and evaluation were based on MD in Sox-10-stained digitalised slides. In total, 86 whole slide images (WSIs) from LM patients were annotated and used as a training set. The test set consisted of 177 slides. The tool was trained to detect the epidermis, measure its length, and determine the MD. A cut-off of ≥30 melanocytes per 0.5 mm of epidermis length was defined as positive. Our AI model automatically recognises the epidermis and measures the MD. The model was trained on nuclear immunohistochemical signals and can also be applied to other nuclear stains, such as PRAME or MITF. The WSI is automatically visualised by a three-colour heat map with a subdivision into low, borderline, and high melanocyte density. The cut-offs can be adjusted individually. Compared to manually counted ground truth MD, the AI model achieved high sensitivity (87.84%), specificity (72.82%), and accuracy (79.10%), and an area under the curve (AUC) of 0.818 in the test set. This automated tool can assist (dermato) pathologists by providing a quick overview of the WSI at first glance and making the time-consuming assessment of resection margins more efficient and more reproducible. The AI model can provide significant benefits in the daily routine workflow. Full article

Mobile Mapping Systems (MMSs) have emerged as indispensable instruments for producing high-precision road maps in recent years. Despite incorporating modern devices, their efficacy may be influenced by operational variables such as vehicle speed or the type of GNSS (Global Navigation Satellite System) differential correction employed. This study assesses the impact of varying vehicle speeds and differential correction settings on the accuracy of point grids acquired with an MMS on a two-lane rural road. The experiment was performed across a 7 km distance, incorporating two speeds (40 and 60 km/h) and two travel directions. Three correction methodologies were examined: a proximate local base (MBS), a network station solution of the National Geographic Institute (NET), and virtual reference stations (VRSs). The methodology encompassed normality analysis, descriptive statistics, mean comparisons, one- and two-factor analysis of variance (ANOVA), and the computation of the root mean square error (RMSE) as a measure of accuracy. The findings indicate that horizontal discrepancies remain steady and unaffected by the correction technique; however, notable changes are seen in the vertical component, with the NET option proving to be the most effective. The acquisition rate is the primary determinant, exacerbating errors at 60 km/h. In conclusion, the dependability of MMS surveys is contingent upon the correction approach and operational conditions, and it is advisable to sustain moderate speeds to guarantee precise three-dimensional models. Full article

Traffic signs exposed to long-term outdoor conditions frequently exhibit deformation, inclination, or other forms of physical damage, highlighting the need for timely and reliable anomaly assessment to support road safety management. While point-cloud data provide accurate three-dimensional geometric information, their sparse distribution and lack of appearance cues make traffic sign extraction challenging in complex environments. High-resolution sequential video-log images captured from multiple viewpoints offer complementary advantages by providing rich color and texture information. In this study, we propose an integrated traffic sign detection and assessment framework that combines video-log images and mobile-mapping point clouds to enhance both accuracy and robustness. A dedicated YOLO-SIGN network is developed to perform precise detection and multi-view association of traffic signs across sequential images. Guided by these detections, a frustum-based point-cloud extraction strategy with seed-point density growing is introduced to efficiently isolate traffic sign panels and supporting poles. The extracted structures are then used for geometric parameterization and damage assessment, including inclination, deformation, and rotation. Experiments on 35 simulated scenes and nine real-world road scenarios demonstrate that the proposed method can reliably extract and evaluate traffic sign conditions in diverse environments. Furthermore, the YOLO-SIGN network achieves a localization precision of 91.16% and a classification mAP of 84.64%, outperforming YOLOv10s by 1.7% and 8.7%, respectively, while maintaining a reduced number of parameters. These results confirm the effectiveness and practical value of the proposed framework for large-scale traffic sign monitoring. Full article

Deep learning has become a powerful tool for diagnosing pests and plant diseases, although conventional convolutional neural networks (CNNs) generally suffer from limited interpretability and suboptimal focus on important image features. This study examines the integration of attention mechanisms into two prevalent CNN architectures—ResNet50 and MobileNetV2—to improve the interpretability and classification of diseases impacting date palm trees. Four attention modules—Squeeze-and-Excitation (SE), Efficient Channel Attention (ECA), Soft Attention, and the Convolutional Block Attention Module (CBAM)—were systematically integrated into ResNet50 and MobileNetV2 and assessed on the Palm Leaves dataset. Using transfer learning, the models were trained and evaluated through accuracy, F1-score, Grad-CAM visualizations, and quantitative metrics such as entropy and Attention Focus Scores. Analysis was also performed on the model’s complexity, including parameters and FLOPs. To confirm generalization, we tested the improved models on field data that was not part of the dataset used for learning. The experimental results demonstrated that the integration of attention mechanisms substantially improved both predictive accuracy and interpretability across all evaluated architectures. For MobileNetV2, the best performance and the most compact attention maps were obtained with SE and ECA (reaching 91%), while Soft Attention improved accuracy but produced broader, less concentrated activation patterns. For ResNet50, SE achieved the most focused and symptom-specific heatmaps, whereas CBAM reached the highest classification accuracy (up to 90.4%) but generated more spatially diffuse Grad-CAM activations. Overall, these findings demonstrate that attention-enhanced CNNs can provide accurate, interpretable, and robust detection of palm tree diseases and pests under real-world agricultural conditions. Full article

This study evaluated the applicability of the dual drainage model, Hyper Connected–Solution for Urban Flood (HC-SURF), for real-time urban flood forecasting. The model was applied to the extreme rainfall event of August 2022 in the Sillim and Daerim drainage basins in Seoul. Its accuracy and computational efficiency were quantitatively compared with those of two widely used commercial models, the Personal Computer Storm Water Management Model (PC-SWMM) and InfoWorks Integrated Catchment Modelling (ICM). Accuracy was assessed by measuring spatial agreement with observed inundation trace maps using binary indicators, including the Critical Success Index (CSI), Probability of Detection (POD), and False Alarm Ratio (FAR). Computational efficiency was evaluated by comparing simulation times under identical conditions. In terms of accuracy against observations, HC-SURF achieved CSI values ranging from 0.26 to 0.45, with POD values from 0.37 to 0.81 and FAR values from 0.49 to 0.53 across the two basins. In inter-model comparisons, the model showed high hydraulic consistency, demonstrating CSI values between 0.72 and 0.88, POD between 0.82 and 0.99, and FAR between 0.08 and 0.15. In terms of computational efficiency, HC-SURF reduced calculation times by approximately 9% and 44% compared with InfoWorks ICM and PC-SWMM, respectively, for a 48 h simulation. The model also completed a 6 h rainfall simulation in approximately 8 min, meeting the lead time requirements for rapid urban flood forecasting. Overall, these findings show that HC-SURF effectively balances simulation accuracy with computational efficiency, demonstrating its suitability for real-time urban flood forecasting. Full article

In order to reduce the risk of collapse disasters during tunnel construction in mountainous areas and to make full use of the available data, a collapse risk assessment model for highway tunnel construction was established based on the WOA–XGBOOST algorithm. Three major categories of tunnel construction risk, namely engineering geological factors, survey and design factors, and construction management factors, were selected as the first-level indicators, and 14 secondary indicators were further specified as the input variables of the collapse risk assessment model for tunnel construction. The confusion matrix and accuracy metrics were employed to evaluate the training and prediction performance of the risk assessment model on both the training set and the test set. The results show that subjective weights derived from the G1 method were integrated with objective weights generated by the WOA–XGBOOST algorithm. A game-theory-based weight integration strategy was then applied to optimize the combined weights, effectively mitigating the biases inherent in single-method weighting approaches. Risk quantification was systematically conducted using a cloud model, while spatial risk distribution patterns were visualized through graphical cloud-mapping techniques. After completion of model training, the proposed model achieved a high accuracy of over 99% on the training set and around 95% on the held-out test set based on an available dataset of 100 collapse-prone tunnel construction sections. Case-based verification further suggests that, in the studied collapse scenarios, the predicted risk levels are generally consistent with the actual engineering risks, indicating that the model is a promising tool for assisting tunnel construction risk assessment under similar conditions. The research outcomes provide an efficient and reliable approach for assessing risks in tunnel construction, thereby offering a scientific basis for engineering decision-making processes. Full article

Urban green space is essential for ecological functions, environmental quality, and human well-being, yet campus expansion can reduce vegetated areas. This study assessed UGS dynamics at Universitas Padjadjaran’s Jatinangor campus from 2015 to 2025 and evaluated an object-based machine learning approach for fine-scale land cover mapping. High-resolution WorldView-2, WorldView-3, and Legion-03 imagery were pan-sharpened, geometrically corrected, normalized, and used to compute NDVI and NDWI indices. Object-based image analysis segmented the imagery into homogeneous objects, followed by random forest classification into six land cover classes; UGS was derived from dense and sparse vegetation. Accuracy assessment included confusion matrices, overall accuracy 0.810–0.860, kappa coefficients 0.747–0.826, weighted F1 scores 0.807–0.860, and validation with 43 field points. The total UGS increased from 68.89% to 74.69%, bare land decreased from 13.49% to 5.81%, and building areas moderately increased from 10.36% to 11.52%. The maps captured vegetated and developed zones accurately, demonstrating the reliability of the classification approach. These findings indicate that campus expansion has been managed without compromising ecological integrity, providing spatially explicit, reliable data to inform sustainable campus planning and support green campus initiatives. Full article