Search Results (887)

Urban parks are often regarded as carbon sinks, yet their net carbon performance depends on the balance between vegetation carbon uptake and maintenance-related emissions, as well as the accurate representation of within-park spatial heterogeneity. This study used backpack LiDAR, field vegetation surveys, and maintenance inventories to quantify annual carbon sequestration, maintenance emissions, and net carbon budget in 44 plots covering nine biotope types across 16 parks in central Xianyang, China. A four-level biotope classification incorporating canopy openness, ground cover, tree composition, and vertical stratification was applied to link LiDAR-derived three-dimensional structure with ecological-unit-level carbon accounting. Carbon sequestration and net carbon budget differed significantly among biotopes, whereas maintenance emissions did not. Closed broadleaved single-layer forest showed the highest carbon sequestration density (0.772 kg C m⁻²), while hard-surfaced partly closed broadleaved single-layer forest showed the lowest value (0.132 kg C m⁻²). Closed woody biotopes functioned as strong carbon sinks, partly closed biotopes as weak sinks, and the partly open short-grass biotope was the only carbon source. Three-dimensional green volume density was the strongest positive predictor of net carbon budget (β = 0.417, p = 0.032), followed by stem density (β = 0.276, p = 0.048), whereas irrigation-related emissions showed a significant negative coefficient (β = −0.276, p = 0.021). Carbon sequestration explained more variation in net carbon budget than maintenance emissions (adjusted R² = 0.409 vs. 0.134). These findings suggest that backpack LiDAR can support fine-scale identification of priority carbon-sink units in urban parks and that low-carbon park management should prioritize three-dimensional woody vegetation structure while reducing high-input irrigation where feasible. Full article

Woody vegetation growing outside officially designated forest land represents a significant but poorly quantified resource in many countries, where institutional and methodological limitations hinder its systematic accounting. This study develops and applies a multi-stage remote sensing-based framework to identify and characterize forest-eligible areas (FEAs) in Lithuania by integrating airborne LiDAR, Sentinel-2 time series, historical orthophotos, and national geospatial datasets. The workflow combines (i) LiDAR-derived canopy height model generation and object-based segmentation, (ii) rule-based aggregation of vegetation segments according to legal forest criteria, (iii) multi-index Sentinel-2 change detection to exclude recent disturbances, and (iv) deep learning-based classification of historical orthophotos to assess stand age. Three detection approaches were evaluated—LiDAR-based, land parcel identification system (LPIS)-based, and their combination. A total of 111,754.4 ha of FEAs were identified outside official forest land, of which 76,204.6 ha meet the minimum age criterion for classification as forest land under national legislation. The designation of these areas as forest land would increase national forest cover from 33.9% to 35.0%. The LiDAR-based approach achieved the highest overall accuracy after dataset refinement (91.5%), while the combined approach yielded the highest precision (97.1%). Accuracy improved notably when reference points affected by definitional conflicts and temporal inconsistencies were excluded, indicating that apparent detection errors were largely attributable to reference data limitations rather than algorithmic failure. The proposed framework offers a scalable solution for wall-to-wall identification and monitoring of unregistered forest resources, with direct applications for national forest inventories and LULUCF reporting. Full article

Rangelands occupy over half of the Earth’s terrestrial surface and play an important role in supporting biodiversity and livelihoods. However, widespread degradation—particularly in arid and semi-arid regions—has compromised their ecological function. Traditional monitoring approaches that rely on vegetation cover metrics from optical satellite imagery fail to capture the three-dimensional structure of vegetation, which is critical for assessing ecosystem condition and guiding restoration and management efforts. This study demonstrates the application of high-density airborne LiDAR (ALS) data (~15–20 points/m²) to identify and map vegetation structural types across 370,000 hectares of semi-arid rangelands in Western Australia. Using an unsupervised fuzzy c-means clustering algorithm on seven minimally correlated ALS-derived structural metrics, we identified eight statistically distinct vegetation structural classes. The resulting structural map revealed spatial heterogeneity in vegetation structure, including in areas with similar vegetation cover, with high confidence in structural attribution in 74.5% of the study area. The rangeland-specific structural classes developed in this study, which incorporate measures of classification certainty, offer a robust framework for vegetation structural mapping in field data-scarce environments. This framework can support ecological condition assessments and provide a basis for rangeland management and restoration planning. Full article

High-rise LiDAR scanning produces vertically sparse point clouds where upper-layer defects are hardest to detect due to inverse-square ranging law (1/r²) density gradients, noise contamination, and complex geometries. This paper presents PC-TowerNet, a physics-aware AI pipeline that achieves state-of-the-art reconstruction through sequential modules: (1) 50D geometric feature classification outperforming CloudCompare SOR (100% accuracy vs. 91.3% retention); (2) Physics-Constrained VAE (PC-VAE) recovering 28.7 ± 2.1% upper density vs. 8.3 ± 1.7% standard VAE; (3) multi-modal PointNet++/GNN/Transformer fusion; and (4) Bayesian uncertainty maps (ECE = 0.042 ± 0.008). Synthetic tower evaluation (10 × 5 seeds) demonstrates 48.9% surface smoothness improvement and 38.2% volume error reduction over tuned RANSAC baselines, with clear paths to real-data validation. Full article

Crop attribute detection, as a key component of intelligent agricultural harvesting machinery, plays a crucial role in harvesting efficiency, loss reduction, and autonomous operation control. Compared with existing reviews on artificial intelligence and sensing technologies in agriculture, this review focuses on crop attribute detection scenarios oriented toward the intelligent decision-making and control requirements of agricultural harvesting machinery. It mainly analyzes crop attributes that affect harvesting operations, as well as the sensors and algorithms involved in detecting these attributes, and further clarifies the relationship between detection methods and control decisions in agricultural harvesting machinery. For grain crops, the key attributes relevant to harvesting operations include plant height, plant density, spike number, crop lodging, canopy structure, and crop position. For fruit and vegetable crops, the key attributes relevant to harvesting operations include maturity, position, and quality. From the perspectives of multi-source data acquisition, data analysis, and attribute detection algorithms, the key technologies in the field of crop attribute detection are systematically summarized and analyzed, including sensors used in crop attribute detection, such as RGB, spectral, near-infrared, and LiDAR sensors, as well as data analysis and recognition approaches, such as image classification, object detection, and point cloud analysis. The complexity of field environments and the dynamics of machine operation are analyzed, highlighting the technical bottlenecks of current detection systems in environmental adaptability, real-time responsiveness, and resistance to interference. To address these challenges, feasible optimization directions were proposed, including multi-sensor fusion, weakly supervised learning, and few-shot learning. This review aims to provide systematic references and theoretical support for the coordinated development of crop detection and control decision-making in intelligent agricultural harvesting systems. Full article

Although deep learning has been successfully used to detect landslide hazards in recent years, existing methods still face challenges due to the variety of landslide characteristics in different terrains and topographies. This study proposes a new framework for landslide detection by comparing various YOLO models. It employs deformable convolutional modules combined with GhostConv modules to enhance feature extraction for landslide targets. The framework uses a structured IoU loss function to optimize the alignment of actual and predicted frames in a directional sense. Additionally, it introduces the CoordAtt attention mechanism to accelerate model convergence and improve training efficiency. The experimental results demonstrate that the enhanced YOLO model (CDCS-YOLO), incorporating four key enhancement modules (Coordinate Attention, Deformable Convolutional Networks, the C3 Module/CSP Architecture and SIoU Loss), achieved a maximum mAP of 96.6%, an accuracy of 96.1%, and a frame rate of 142.6 FPS. Notably, it performed exceptionally well in soil landslide detection, achieving an average detection accuracy surpassing 90%. Based on the experimental results, we explored a morphological landslide classification method further as well as a multi-source differential monitoring strategy integrating UAV imagery, field surveys, ground-based LiDAR data, rainfall information and deformation indicators. The proposed method outperforms the baseline approach and is a promising solution for detecting landslides and geological hazards in Xinjiang. Full article

Remote sensing enables building-level flood vulnerability assessment without field surveys, yet existing approaches require site-specific calibration or produce categorical outputs without physical interpretability. We present the Global Flood Vulnerability Model (GFVM), integrating six remotely sensed components (elevation, slope, topographic position index, distance to water, building height, and basement depth) through geographic context classification to quantify vulnerability from terrain and structural characteristics across coastal, fluvial, and pluvial settings. Building heights are extracted primarily from the Global Building Atlas, with gaps filled using a ConvNeXt neural network trained on high-resolution Light Detection and Ranging (LiDAR) ground truth from four cities (within-city MAE 1.35–1.91 m, cross-city MAE 2.05–3.47 m). Terrain metrics are derived from a combination of hierarchical digital elevation models (DEM) (USGS 3DEP 10 m, AHN LiDAR 0.5 m, UK Environment Agency DTM 1 m, Australia 5 m) and global datasets (NASADEM 30 m, Copernicus GLO-30). Hydrographic networks are sourced from OpenStreetMap and Natural Earth. Implementation through Google Earth Engine requires only coordinates as input, returning a five-level vulnerability index with multi-hazard decomposition (fluvial, coastal, pluvial) and SHapley Additive exPlanations (SHAP)-based attribution identifying dominant drivers. Validation across 183 independent locations in Germany, UK, and USA demonstrates robust performance: Area Under Curve 0.855 for separating flooded from non-flooded sites, weighted Cohen’s kappa 0.493 across regulatory zones, and Spearman ρ 0.746 against Federal Emergency Management Agency (FEMA) classifications. Sensitivity analysis across 625 parameter configurations confirms stability, and DEM resolution experiments show that global 30 m elevation data produces category reclassification in only 5.3–8.6% of locations compared to high-resolution sources. Application to the 2024 Kazakhstan floods identifies 118 high-vulnerability locations across 581 assessment points, with vulnerability patterns matching documented inundation. GFVM advances remote sensing applications for disaster risk assessment by demonstrating that multi-source geospatial data fusion enables building-level vulnerability screening without local calibration or field surveys. Full article

High-latitude coastal regions are highly sensitive to climate change, yet their geomorphology is obscured by sea ice, landfast ice and seasonal snow, restricting the applicability of optical remote sensing for fine coastal classification. To address this limitation, we develop an interpretable coastal classification framework integrating ICESat-2 photon-counting LiDAR and explainable machine learning. Multi-dimensional morphometric features describing cross-shore geometry, vertical relief and local slope variability are extracted from ICESat-2 ATL03 along-track profiles to train a CatBoost classifier, with five-fold cross-validation and sample weighting to mitigate class imbalance. Introducing SHAP-based interpretability into ICESat-2-driven coastal geomorphic classification enables the identification of morphometric controls on coastal-type differentiation. Validated in the Bering Sea with 447 profiles and a 75%/25% stratified split, the framework achieved an overall accuracy of 86.6%, a macro-average recall of 89.4% and a Kappa coefficient of 0.84. SHAP analysis identifies that coastal width is the most influential feature for model-based classification of coastal geomorphic types, while slope and local steepness variability serve as important predictive indicators for distinguishing rocky and sedimentary coasts. This framework links data-driven classification to geomorphic processes and provides a potentially generalisable approach for fine-scale coastal mapping in high-latitude environments. Full article

Cross-domain semantic classification of 3D point clouds remains challenging due to strong domain shifts between heterogeneous sensing modalities. Most existing classification frameworks are domain-specific, limiting their use in integrated land–water mapping applications. This study evaluates the transferability of classical geometric machine learning classifiers between terrestrial and underwater point cloud domains without target-domain retraining. Experiments were conducted using terrestrial data acquired with a Leica BLK360 terrestrial laser scanner (TLS) and underwater point clouds collected with a Blueview BV5000 mechanical scanning sonar (MSS). Two dimensionality-based frameworks, CANUPO–Support Vector Machine (SVM) and 3DMASC–Random Forest (RF), were implemented in CloudCompare and assessed under intra-domain and cross-domain configurations. Strong intra-domain performance was achieved, with terrestrial–terrestrial accuracies of 0.99 for CANUPO–SVM and 0.97 for 3DMASC. In underwater evaluation, CANUPO maintained high accuracy (0.97), whereas 3DMASC decreased to 0.86 due to increased variability in the submerged data. Under cross-domain transfer, CANUPO achieved 0.93 accuracy for terrestrial-to-underwater and 0.89 for underwater-to-terrestrial classification, while 3DMASC demonstrated stable generalisation with 0.95 accuracy in both directions. Overall, dimensionality-based geometric descriptors capture stable structural cues across sensing environments, providing an interpretable and efficient pathway for applications such as hydrographic surveying, coastal monitoring, and underwater search-and-rescue detection. Future work will extend validation to larger datasets and explore domain adaptation strategies to further reduce cross-modality domain shift. Full article

The water–land classification is fundamental for shoreline extraction and coastal habitat mapping, which is the basis of a comprehensive assessment and ecosystem-based coastal zone management. This study aims to separate water and land for coastal zones by taking advantage of both high-resolution satellite imagery and airborne lidar point clouds. Considering physical principles of optical remote sensing and lidar, we developed a prior knowledge-based localization classification approach that eliminates the need for collecting training sets and handling temporal differences across multiple data sources. Our approach first created the initial classification using the WorldView-2 (WV2) Normalized Difference Water Index. Then, the Connected Components Labeling algorithm was used to create a non-overlapping partition of the working area. The third step involved processing the water blocks using prior land cover knowledge. Finally, we used lidar point clouds to refine the initial water blocks and their neighboring areas. This classification approach showed promising results along Matagorda Bay, Texas, an approximately 2449 km² area that is covered by 26 WV2 images and 1568 lidar tiles. Full article

The Yellow River Basin contains abundant coal resources; however, its ecological environment is inherently fragile, and vegetation degradation has been further intensified by extensive mining activities. Accurate classification of individual tree species in mining-affected areas is therefore essential for assessing ecological conditions and establishing a scientific foundation for targeted restoration and sustainable management. To address this need, an evaluated machine learning framework was developed and evaluated for individual tree species classification in a coal mining area of the Yellow River Basin using integrated unmanned aerial vehicle (UAV) data. A comprehensive feature set was constructed by extracting 278 attributes per tree. These attributes included 224 spectral bands and 29 hyperspectral indices derived from hyperspectral imagery, 24 textural metrics obtained from RGB orthophotos, and one canopy height feature generated from a LiDAR-derived model. Based on ground-truth data from 1095 individual trees, seven machine learning algorithms were trained and systematically compared: Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree (DT), Gradient Boosting (GB), Logistic Regression (LR), and XGBoost. Statistical significance testing using 5 × 5 repeated cross-validation, together with the Friedman test and post hoc Nemenyi test, and additional model stability analysis consistently identified XGBoost as the optimal classifier. On an independent test set, XGBoost achieved high accuracy (Overall Accuracy = 0.897, Kappa = 0.811) with an efficient training time of 2.36 s. Further analysis demonstrated the critical and complementary roles of hyperspectral and structural features in species discrimination. The optimized model was subsequently applied to generate a detailed wall-to-wall tree species map across the entire mining area. Overall, this study presents a statistically informed comparison of classifiers for multi-source feature-based species discrimination and delivers an evaluated and practical pipeline for effective vegetation monitoring. The proposed framework provides a scientific tool for assessing and managing ecological recovery in complex mining environments, particularly within ecologically sensitive regions such as the Yellow River Basin. Full article

Accurate identification of tree species is important for forest management, biodiversity studies, and precision forestry. Near-range LiDAR point clouds provide detailed three-dimensional information about individual trees. However, the complex structure of the point clouds and the unbalanced distribution of species make automatic classification difficult. To address these issues, this study presents a Transformer model with geometric enhancement. The model combines local geometric features and global attention to improve species recognition in forest environments. It uses geometric information with biological meaning, including point cloud normals, local density, vertical structure, and growth direction. A focal loss with class balance is also introduced to reduce the impact of species distributions with long tails. Experiments on the ForSpecial20K dataset show that the proposed method performs better than representative models based on convolution, graph methods, and Transformer architectures. It achieves higher overall accuracy (78.20%), higher mean class accuracy (73.48%), and a higher Macro-F1 score (73.21%). Results from confusion matrices and visual analysis of similar species further verify the effectiveness of the geometric features and the loss design. These results suggest that modeling structural information of forests helps improve robustness and generalization. The proposed method offers a practical solution for tree-level species mapping, fusion of LiDAR data from multiple sources, and fine-scale forest inventory. It also shows the value of combining high-resolution LiDAR data with deep learning for forestry applications. Full article

Many place recognition approaches, which identify previously visited places or locations by matching current sensory data, such as 2D RGB images and 3D point clouds, have been proposed to achieve accurate and robust localization and loop closure detection in global positioning system (GPS)-denied environments. Since visual place recognition (VPR) methods that rely on images captured by camera sensors are highly sensitive to variations in appearance, including changes in lighting, surface color, and shadows, they can lead to poor place recognition accuracy. In contrast, light detection and ranging (LiDAR)-based place recognition (LPR) approaches based on 3D point cloud data that captures the shape and geometric structure of the environment are robust to changes in place appearance and can therefore provide more reliable place recognition results than VPR methods. This work presents an indoor LPR method called PointNetVLAD-based indoor pedestrian localization (PIPL). PIPL is a deep network model that uses PointNetVLAD to learn to extract global descriptors from 3D LiDAR point cloud data. PIPL can recognize places previously visited by a pedestrian using point clouds captured by a low-cost LiDAR sensor on a smartphone in small-scale indoor environments, while PointNetVLAD performs place recognition for vehicles using high-cost LiDAR, GPS, and inertial measurement unit (IMU) sensors in large-scale outdoor areas. For place recognition on 3D point cloud reference maps generated from LiDAR scans, PointNetVLAD exploits the universal transverse mercator (UTM) coordinate system based on GPS and IMU measurements, whereas PIPL uses a virtual coordinate system designed in this study due to the unavailability of GPS indoors. In experiments conducted in campus buildings, PIPL shows significant advantages over NetVLAD (known as a convolutional neural network (CNN)-based VPR method). Particularly in indoor environments with repetitive scenes where geometric structures are preserved and image-based appearance features are sparse or unclear, PIPL achieved $39\%$ higher top-1 accuracy and $10\%$ higher top-3 accuracy compared to NetVLAD. Furthermore, PIPL achieved place recognition accuracy comparable to NetVLAD even with a small number of points in a 3D point cloud and outperformed NetVLAD even with a smaller model training dataset. The experimental results also indicate that PIPL requires over $76\%$ less place retrieval time than NetVLAD while maintaining robust place classification performance. Full article

Accurate extraction of impervious surface areas (ISA) is essential for urban environmental monitoring, yet severe spectral confusion among complex urban land-cover types limits the performance of classifications based solely on optical imagery. To address this issue within a localized context, this study proposes a multi-source framework integrating UAV-based LiDAR (UAV-LiDAR) and high-resolution visible imagery for fine-scale ISA extraction. An improved segmentation optimization strategy, termed EGS-Optimizer, is developed to enhance boundary delineation within the object-based image analysis (OBIA) framework by coupling edge detection with global segmentation quality evaluation. A comprehensive feature set including spectral, index, texture, geometric, and terrain features is constructed, and Shapley Additive Explanations (SHAP) is applied to select the most informative variables while reducing dimensionality. The proposed framework is validated in a typical 1.45 km² built-up area in Deyang City, Sichuan Province. Experimental results demonstrate that, within this specific study area, multi-source data fusion improves classification accuracy by 3.59–5.79% compared with single-source data, while feature selection reduces the feature dimension from 45 to 21. Among the evaluated classifiers, the random forest (RF) model achieves the highest performance, with an overall accuracy of 97.24% (Kappa = 0.96). While the high accuracy highlights the efficacy of synergizing spectral and structural information for micro-landscape mapping, these findings are constrained to the demonstrated fine-scale local environment. The results provide an effective, interpretable solution for detailed neighborhood-level ISA mapping, though further validation is required before the framework can be generalized to larger or more heterogeneous urban scenarios. Full article

Diameter at Breast Height (DBH) is a key parameter in forest measurement. However, existing research has mostly focused on improving the accuracy of individual technologies, lacking a systematic synthesis of the evolutionary logic of measurement techniques and a standardized selection framework for forestry applications. To this end, this paper constructs a multi-level classification framework based on measurement platforms and technical principles, establishes for the first time a five-dimensional comprehensive evaluation system (covering accuracy, efficiency, cost, environmental adaptability, and automation) along with a hierarchical technology decision tree, and systematically analyzes the application logic of multi-source fusion technologies across three levels: ground-based, near-ground mobile, and aerial. The review indicates that traditional contact-based measurement has limited efficiency; modern remote sensing technologies (photogrammetry and LiDAR) offer significant advantages in automation and accuracy, but still face challenges such as high equipment costs, complex data processing, and poor environmental adaptability. Multi-source fusion and machine learning are key methods to overcome the limitations of single sensors and improve the robustness of DBH estimation. Finally, it is anticipated that with decreasing sensor costs and the advancement of intelligent algorithms, DBH measurement will continue to evolve toward automation, intelligence, and engineering practicality, providing technical support for large-scale, long-term, and repeatable forest monitoring. Full article