International journal of applied earth observation and geoinformation : ITC journal最新文献

{"title":"Mangrove-Net: A multi-stage attention fusion and class token network for mangrove species mapping using lidar point clouds","authors":"Dezhi Wang ,&nbsp;Ziyang Wu ,&nbsp;Penghua Qiu ,&nbsp;Simeng Wang ,&nbsp;Bo Wan ,&nbsp;Zongzhu Chen ,&nbsp;Sheng Nie ,&nbsp;Yihui Zhang","doi":"10.1016/j.jag.2026.105166","DOIUrl":"10.1016/j.jag.2026.105166","url":null,"abstract":"<div><div>Mapping mangrove species is crucial for mangrove biodiversity protection and ecological restoration. Existing mangrove classification methods mostly rely on the height and intensity metrics of rasterized lidar point clouds, which lead to the partial loss of fine three-dimensional (3D) structural information, and are difficult to distinguish among small inter-class differences, complex structures (dense canopy and aerial roots), and uneven mangrove stands. To directly capture the species community structure of mangroves from a 3D perspective, this study develops a point-wise deep learning species classification method specifically for mangroves—called Mangrove Multi-stage Attention Fusion and Class Token Network (Mangrove-Net), using UAV lidar point clouds and RGB imagery fusion data. Mangrove-Net uses a multi-stage attention fusion and class token mechanism to balance local geometric representation with global semantic perception, effectively capturing representative features among mangrove species. To address sample-imbalanced classification tasks, Mangrove-Net introduces a weighted cross-entropy loss function. We tested the proposed method on a typical mangrove forest dataset (from the Hainan Qinglangang Provincial Nature Reserve, China) and on the open dataset ModelNet40. The results indicated that Mangrove-Net showed superior performance compared to six state-of-the-art point-based deep learning methods (such as PointNet++, Point Transformer, PointMLP, and GPSFormer) in mapping fine-grained mangrove species, with a minimum improvement of 6.60% and 7.71% in overall accuracy (OA) and mean category accuracy (mAcc), respectively. Multi-scale visualization (sample, local, and study area) showed that the mangrove species maps generated by our proposed method had higher spatial continuity and ecological consistency. Evaluated on the ModelNet40 dataset, Mangrove-Net also surpassed those baseline methods, further demonstrating the effectiveness and robustness of the method. This study provides a reliable point-wise deep learning method specifically for mangrove species classification to support various tasks in fine-grained precision mangrove forestry.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105166"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147278286","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Widespread biophysical cooling effects due to post-fire greening","authors":"Huipeng Xi ,&nbsp;Qunming Wang ,&nbsp;Yuelong Xiao ,&nbsp;Ru Guo ,&nbsp;Xiaohua Tong ,&nbsp;Peter M. Atkinson","doi":"10.1016/j.jag.2026.105211","DOIUrl":"10.1016/j.jag.2026.105211","url":null,"abstract":"<div><div>Wildfires and greening are two important biophysical processes that influence land surface-climate feedback patterns. However, the impact of post-fire greening, which primarily reflects canopy structural recovery, on land surface temperature (LST) remains uncertain, particularly at the daily scale, as this temporal resolution allows for a clearer observation of how fire seasonality and vegetation regrowth influence short-term land surface energy dynamics. In this research, using satellite sensor observations covering the globe from 2004 to 2019, we found that the median post-fire recovery time of leaf area index (LAI) was 479.5 days. Most forests exhibited faster canopy LAI recovery than low-stature herbaceous vegetation, likely due to differences in burn severity and affected plant structures. Fire seasonality also shaped LAI recovery patterns: dry-season and spring fires led to quicker regrowth, with the shortest recovery in temperate spring fires (173.7 days), while wet-season and summer fires showed delayed recovery, especially in cold zones where summer fires need 425.5 days to recovery. During post-fire greening, the annual cycle of LAI recovery caused an average cooling effect of −0.04 K d^-1, due to the strong evapotranspiration-climate negative feedback. However, seasonal effects varied: summer fires in temperate and cold zones led to cooling, with the strongest warming observed after temperate winter fires (up to 0.104 K d⁻¹). Furthermore, we observed a widespread decrease in carbon use efficiency during post-fire LAI recovery, which means that the recovery rate of ecosystem carbon sinks may not be synchronized with the rate of vegetation greening. By distinguishing between structural and functional recovery, we found that early evapotranspiration-driven cooling during structural recovery may not persist throughout ecosystem functional recovery. This study enhances our understanding of the global biophysical climate effects of post-fire greening in the context of the earth’s land surface-climate feedback, and reveals precise changes in the component parts of this feedback effect.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105211"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147360587","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Constrained planar primitive segmentation for building component extraction from airborne LiDAR point clouds","authors":"Hongxin Yang ,&nbsp;Zhipeng Luo ,&nbsp;Dedong Zhang ,&nbsp;Jonathan Li","doi":"10.1016/j.jag.2026.105182","DOIUrl":"10.1016/j.jag.2026.105182","url":null,"abstract":"<div><div>Accurately segmenting planar primitives from airborne LiDAR point clouds is crucial for urban planning applications and three-dimensional (3D) building reconstruction. However, existing approaches often exhibit limited extensibility and diminished robustness when segmenting buildings with diverse architectural styles. To address this issue, this paper proposes an enhanced constrained k-Plane Clustering (kPC) method that segments building point clouds into distinct planar primitives. The kPC algorithm formulates the segmentation task as a mixed-integer non-convex optimization problem incorporating three geometric constraints: point-to-plane distance minimization, cluster-center proximity enforcement, and directional regularization. This problem is solved via an alternating minimization strategy, which iteratively updates cluster assignments and plane parameters using Singular Value Decomposition (SVD) until convergence is reached. The proposed constrained kPC method provides two key advantages. First, it effectively mitigates the infinite extensibility of fitted planes. This issue is common in conventional kPC methods, and its mitigation directly addresses a major source of suboptimal segmentation performance. Second, the framework demonstrates robustness to variations in the optimization objective’s coefficients while maintaining consistent performance. Extensive experiments on both synthetic and real-world multi-style building datasets demonstrate that the proposed method achieves superior segmentation accuracy and outperforms state-of-the-art approaches in both qualitative and quantitative evaluations. Furthermore, the performance of the proposed method is robust to variations in its parameters, maintaining effectiveness across a wide range of values.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105182"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147360593","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Individual sapling-level leaf chlorophyll content estimation in Ginkgo biloba with weed background: A novel deep-learning approach coupling APUNet-YOLO with PROSAIL-DNN","authors":"Lin Su ,&nbsp;Kai Zhou ,&nbsp;Nicholas C. Coops ,&nbsp;Justin Morgenroth ,&nbsp;Liam A.K. Irwin ,&nbsp;Donald White ,&nbsp;Jimei Han ,&nbsp;Xin Shen ,&nbsp;Lin Cao","doi":"10.1016/j.jag.2026.105194","DOIUrl":"10.1016/j.jag.2026.105194","url":null,"abstract":"<div><div>Accurate estimation of chlorophyll content helps understand the physiological health of saplings. Traditional field measurement methods limit large-scale applications, while remote sensing monitoring methods, though improving efficiency, still face challenges in data analysis and processing. A key challenge is extracting the canopy of individual trees and estimating chlorophyll content amid interference from weed vegetation. To overcome this challenge, this study develops a novel coupled deep learning approach. First, an Attention and Pyramid U-Net (APUNet) was proposed to improve the segmentation accuracy of Ginkgo seedlings crown boundaries in multispectral drone images. Second, the hyper-parameter combination of the You Only Look Once (YOLO) model was optimized to improve sapling detection accuracy under complex conditions with high weed coverage. Next, a sliding window method was developed to reduce duplicate detection rates for detecting all saplings in the entire image, and the intersection of APUNet and detection results was used for segmentation of individual saplings. Finally, a PROSAIL-Deep Neural Network (DNN) model for chlorophyll estimation was developed, and the impact of using APUNet YOLO to remove weed pixels on chlorophyll estimation was investigated, along with how over-coverage and under-coverage during the segmentation process affect the accuracy of chlorophyll estimation. The results show that, compared to U-Net, APUNet improved segmentation performance, with the IoU increasing from 0.697 to 0.736. After hyper-parameter optimization through grid search, the YOLOv8l model achieved an mAP50 of 0.968 and when this model was applied to full-image detection using a sliding window approach, its duplicate detection rate (ddr) decreased by 11.2% compared with the general method. In chlorophyll inversion, the chlorophyll estimation accuracy reached its highest after using APUNet-YOLO to reduce the interference from weed pixels, with an R² of 0.752 and an RMSE of 7.143 μg/cm². Meanwhile, the sensitivity analysis of the impact of segmentation errors on chlorophyll estimation accuracy shows that, compared to under-coverage segmentation, over-inclusion segmentation leads to a greater decline in chlorophyll estimation accuracy. This novel coupled model can effectively monitor the chlorophyll content of Ginkgo saplings in settings with complex backgrounds, providing a practical possibility for the realization of precision forestry.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105194"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147360642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MAC-DETR: Re-parameterized Mambaout for robust object detection in high-resolution remote sensing","authors":"Huanxu Li ,&nbsp;Keke Xu ,&nbsp;Xianglei Liu ,&nbsp;Jianlong Wang ,&nbsp;Guoqing Zhou","doi":"10.1016/j.jag.2026.105169","DOIUrl":"10.1016/j.jag.2026.105169","url":null,"abstract":"<div><div>Accurate and efficient object detection in remote sensing imagery is fundamental to applications such as urban monitoring, infrastructure assessment, and disaster management. However, two persistent challenges remain: (1) small objects occupy only a few pixels in ultra-high-resolution images, making them difficult to detect reliably, and (2) the rapid growth of large-scale satellite data demands computationally efficient methods suitable for practical deployment. Existing lightweight convolutional networks provide real-time efficiency but often fail to capture sufficient discriminative information for small targets, while Transformer-based detectors improve representation power at the cost of high computational complexity. To address these challenges, we propose MAC-DETR, a CNN-Mambaout-Transformer hybrid detection framework designed for remote sensing imagery. MAC-DETR integrates three complementary modules: (1) Mambaout-RepLK, a Mambaout-style gated convolution block that reduces computational overhead while preserving expressive feature extraction, achieving + 2.5% mAP improvement in fine-grained categories on NWPU VHR-10 while reducing parameters by 23%; (2) CROSS-UP, an adaptive upsampling block that enhances multi-scale fusion and improves the detection of small objects—yielding up to + 2.6% mAP gain on multi-scale categories in NWPU VHR-10—without introducing extra complexity; and (3) ASSAIFI, a sparse attention module that strengthens feature interaction to further refine small-object representations. Experiments on four benchmark datasets demonstrate that MAC-DETR achieves 74.1% mAP on DOTA-v1.5, 95.4% on NWPU VHR-10, 94.4% on HRSC2016, and 95.4% on the infrared SIRSTv2 dataset, consistently outperforming both CNN– and Transformer-based baselines. Ablation studies show that the design reduces parameters by 23% and computation by 10–15%, offering a practical balance between accuracy and efficiency. These results highlight MAC-DETR’s effectiveness for large-scale, multi-modal Earth observation applications. The source code will be available at: https://github.com/Keykeykeykeykeykey/MAC-DETR.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105169"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147360851","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Differential land deformation patterns and response mechanism to groundwater recovery in the Beijing-Tianjin-Hebei Plain revealed by reconstructing and clustering of TS-InSAR observations","authors":"Dexin Meng ,&nbsp;Beibei Chen ,&nbsp;Shubo Zhang ,&nbsp;Huili Gong ,&nbsp;Chaofan Zhou ,&nbsp;Sidi Chen ,&nbsp;Kunchao Lei ,&nbsp;Haigang Wang","doi":"10.1016/j.jag.2026.105196","DOIUrl":"10.1016/j.jag.2026.105196","url":null,"abstract":"<div><div>The Beijing-Tianjin-Hebei (BTH) Plain in north China has experienced significant land subsidence due to long-term groundwater overexploitation. Although water diversion project and climate change have improved groundwater conditions overall, the spatiotemporal heterogeneity in land deformation responses to groundwater recovery remains unclear. This study reconstructs and identifies differentiated deformation patterns across the BTH Plain under the new hydrological background, revealing their multi-scale response mechanisms to groundwater change. After mosaicking Time-Series Interferometric Synthetic Aperture Radar (TS-InSAR) deformation fields, we reconstructed the seasonal and long-term components of deformation. A long-term deformation characteristic-constrained K-means clustering method was developed to classify distinct deformation evolution patterns. Coupling these patterns with groundwater-level (GWL) observations to further distinguish the annual lag and long-term delayed response of deformation to GWL changes. From 2016 to 2024, the BTH Plain exhibits clear spatial partitioning of deformation behaviors, categorized as 10.14% continuous subsidence, 18.86% reduced subsidence, 46.09% stable deformation, and 24.91% slight rebound. Uncertainties in the reconstructed deformation reflect short-term fluctuations driven by extreme hydrological or anthropogenic disturbances. Short-time Fourier transform analysis reveals periodic response of deformation to GWL change and annual lag: elastic deformation dominated areas lag GWL by ≤ 2 months, whereas clay-rich persistent subsiding areas lag 4–6 months. The exponential decay model further indicates that the recovery of subsidence funnels is governed by geological structures: funnels intersecting faults show the slowest subsidence decay along their edges, while those without faults display a more uniform edge-to-center recovery pattern.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105196"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147278238","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"GeoAgentic-RAG: A Multi-Agent framework for autonomous geospatial reasoning and visual insight generation with LLM","authors":"Chao Liang ,&nbsp;Yuanzheng Cui ,&nbsp;Run Shi ,&nbsp;Guixiang Zha ,&nbsp;Xin Yin ,&nbsp;Mingzhong Xiao ,&nbsp;Dong Xu ,&nbsp;Xuejun Duan ,&nbsp;Bo Huang","doi":"10.1016/j.jag.2026.105195","DOIUrl":"10.1016/j.jag.2026.105195","url":null,"abstract":"<div><div>Conventional Retrieval-Augmented Generation (RAG) systems have limited effectiveness in geospatial question answering because text-based similarity retrieval cannot adequately represent spatial semantics such as topology and spatial context. To overcome this limitation, we propose GeoAgentic-RAG, a multi-agent framework that enables multimodal large language models (MLLMs) to perform autonomous geospatial reasoning. The framework integrates natural language query parsing, semantic-spatial retrieval, and executable geospatial analysis within a unified, agent-based workflow. Multiple specialized agents collaboratively interpret user queries, retrieve relevant vector and raster datasets from a unified geospatial database, decompose analytical tasks, generate valid spatial logic, and produce interpretable analytical results. We evaluate GeoAgentic-RAG using a benchmark of geospatial retrieval, feature characterization, and spatial relational reasoning tasks in Nanjing and Guangzhou. The proposed framework achieves a pass rate of 85.3% and an answer correctness of 88.3%, outperforming conventional RAG methods and representative code-generation baselines. These results demonstrate that agent-based integration of retrieval and spatial analysis substantially improves the reliability of geospatial question answering and provides a practical framework for the next-generation intelligent GIS applications.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105195"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147278242","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ESCAPE: An ensemble-based self-calibrated autoencoder with physics-informed estimation of high-resolution soil moisture and surface roughness from ALOS-2/PALSAR-2 polarimetric observations","authors":"Jaese Lee ,&nbsp;Haemi Park ,&nbsp;Jungho Im ,&nbsp;Christian Koyama ,&nbsp;Rogelio Ruzcko Tobias ,&nbsp;Takeo Tadono","doi":"10.1016/j.jag.2026.105206","DOIUrl":"10.1016/j.jag.2026.105206","url":null,"abstract":"<div><div>Soil moisture (SM) is a key variable governing land–atmosphere interactions and terrestrial hydrological processes. Recent satellite missions provide global SM observations, among which ALOS-2/PALSAR-2 offers full polarimetric L-band SAR measurements at spatial resolutions as fine as 6 m. Retrieving SM from SAR data requires disentangling scattering contributions from vegetation and surface roughness (e.g., root mean square height; <em>H_rms</em>). Conventional approaches typically rely on collocated in-situ measurements of SM and <em>H_rms</em> for calibration, but such data are often sparse. In addition, the simultaneous estimation of multiple scattering parameters from limited SAR observables can lead to ill-posed inversion and unstable retrievals. To address these challenges, we propose ESCAPE (Ensemble-based Self-Calibrated Autoencoder with Physics-informed Estimation), a self-calibrating framework that integrates polarimetric ALOS-2/PALSAR-2 observations within a 10-member physics-informed neural network (PINN) ensemble. ESCAPE embeds physically based scattering models into an autoencoder architecture and estimates SM and <em>H_rms</em> without using in-situ SM measurements as direct training targets. The ensemble strategy mitigates uncertainty arising from ill-posed physical constraints and gradient-based optimization. Spatial evaluation over Spain and Austria demonstrates robust performance, with a correlation coefficient R = 0.701, an unbiased root mean squared difference (ubRMSD) of 0.089 m³ m⁻³, and a bias of 0.006 m³ m⁻³. Temporal evaluation at an independent site in Japan yields R = 0.568, ubRMSD of 0.06 m³ m⁻³, and a bias of − 0.13 m³ m⁻³. Ensemble analysis further reveals that aggregating multiple PINN realizations improves robustness by reducing the influence of poorly converged members. Compared with conventional approaches trained directly on in-situ data, ESCAPE exhibits improved generalization across heterogeneous environments. These results highlight the potential of ESCAPE as a self-calibrated, satellite-only framework for high-resolution estimation of geophysical parameters, demonstrating the value of combining physical principles with ensemble-based deep learning for SAR remote sensing applications.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105206"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147360591","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automatic alignment of semantic building models with digital city models through multi-level point cloud semantic segmentation and registration using PSO optimization: A case study of the TUM main campus","authors":"Mansour Mehranfar ,&nbsp;Alexander Braun ,&nbsp;André Borrmann ,&nbsp;Medhini Heeramaglore ,&nbsp;Thomas H. Kolbe ,&nbsp;Zhaiyu Chen ,&nbsp;Xiao Xiang Zhu","doi":"10.1016/j.jag.2026.105191","DOIUrl":"10.1016/j.jag.2026.105191","url":null,"abstract":"<div><div>Multi-scale Digital Twins (DTs) of the built environment provide valuable spatial and semantic insights into city assets, supporting urban management, sustainability, and mobility. However, aligning semantic building models with large-scale digital city models remains a significant challenge due to discrepancies in geospatial alignment, spatial resolution, and heterogeneity of data sources. This paper proposes a novel pipeline for the automatic alignment of semantic indoor building models with digital city models. The pipeline comprises two main steps: first, multi-level semantic segmentation of using Artificial Intelligence (AI) models, and second, point cloud registration using the Particle Swarm Optimization (PSO) algorithm, which estimates the transformation parameters required to align the corresponding models. The results of testing the proposed pipeline on real-world data from the Technical University of Munich (TUM) demonstrate the effectiveness of the proposed method in aligning semantic digital models across multiple scales.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105191"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147360592","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Tree-level inversion of leaf chlorophyll content from UAV imagery using a 3D model-based visual factor correction method for adjacency scattering effect","authors":"Shangbo Liu ,&nbsp;Linyuan Li ,&nbsp;Seping Dai ,&nbsp;Yu Su ,&nbsp;Xiaoyao Li ,&nbsp;Yanqiong Li ,&nbsp;Xiankai Lu","doi":"10.1016/j.jag.2026.105200","DOIUrl":"10.1016/j.jag.2026.105200","url":null,"abstract":"<div><div>For tree-level inversion of leaf chlorophyll content (LCC) from UAV imagery, the adjacency scattering effect of individual trees can affect the reflectance of the specific tree crown, thereby increasing the uncertainty and difficulty of LCC inversion. In this study, a novel visual factor (VF) correction framework for such adjacency scattering effect is proposed, which can efficiently quantify the adjacency effects by geometric probabilities instead of computationally intensive 3D radiative transfer modeling (3D-RTM). This framework is realized by extracting the light intersection probability between the target tree crown and the adjacency scenes (Crown, Ground, and Sky) via ray tracing on LiDAR-derived 3D scenes. We constructed a VF-based adjacency effect correction equation to optimize the extracted vegetation indices (VIs) of tree crowns which were utilized for tree-level LCC inversion. The results indicate that most ratio-based vegetation indices that include near-infrared bands, such as NDVI, RVI, MCARI, CIrededge, and GNDVI, increased the accuracy of LCC after calibration (the mean R² of univariate prediction was improved to 0.46), while DVI and EVI were insensitive to the adjacency effect and TVI showed a decrease in accuracy. Notably, VIs after adjacency effect correction demonstrated a remarkable improvement under the multivariate regression algorithm, with the R² increasing by 0.4 compared to that before correction, achieving an accuracy of R² = 0.67. The method overcame the uncertainty of tree-level signals in UAV imagery by eliminating the influence of the adjacency scattering and provided an efficient solution for the accurate inversion of parameters at the individual-tree scale, which is of great application value for the assessment of precise forestry management.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105200"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147360641","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}