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

{"title":"A robust and efficient approach to estimating the age of secondary mangrove forests employing time-series Landsat images and the CCDC model","authors":"Yue Zhang ,&nbsp;Xiaoyan Li ,&nbsp;Rong Zhang ,&nbsp;Lina Cheng ,&nbsp;Mingming Jia ,&nbsp;Chuanpeng Zhao ,&nbsp;Xianxian Guo ,&nbsp;Haihang Zeng ,&nbsp;Wensen Yu ,&nbsp;Qian Shi ,&nbsp;Zongming Wang","doi":"10.1016/j.jag.2025.104789","DOIUrl":"10.1016/j.jag.2025.104789","url":null,"abstract":"<div><div>Secondary mangrove forests are ecosystems that regenerate in areas where original mangrove stands have been degraded or removed as a result of natural disturbances or anthropogenic activities. Compared to mature mangrove forests, secondary stands exhibit enhanced carbon accumulation, increased sediment trapping efficiency, and intensified nitrogen fixation, contributing significantly to coastal eutrophication mitigation. Accurately mapping secondary mangroves and determining their age is essential for sustainable ecosystem management and assessing their services. However, reliably determining mangrove forest age using remote sensing has been hindered by the complex dynamics of intertidal environments. To overcome these challenges, we developed a robust and efficient approach for estimating the age of secondary mangrove forests (ASMF) by integrating Landsat time-series data and the Continuous Change Detection and Classification (CCDC) algorithm. We implemented this method in the Dongzhaigang National Nature Reserve (DNNR), which is the first mangrove nature reserve established in China, achieving a coefficient of determination (R²) of 0.723. Key findings include: (1) the ASMF estimates exhibited high accuracy (R² = 0.723), with optimal performance for forests aged 9–10 years; (2) secondary mangrove forests comprised 47 % (823.87 ha) of the total mangrove area within the DNNR; and (3) younger stands (1–9 years) represented 32 % of all secondary mangrove forests. This approach offers an effective solution for regional-scale mangrove age estimation and provides a critical basis for evaluating the carbon sequestration potential of secondary mangroves in the DNNR.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"143 ","pages":"Article 104789"},"PeriodicalIF":8.6,"publicationDate":"2025-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144860328","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":"CR-CLCD: A cross-regional cropland change detection framework with multi-view domain adaptation for high-resolution satellite imagery","authors":"Zhendong Sun ,&nbsp;Xinyu Wang ,&nbsp;Yanfei Zhong","doi":"10.1016/j.jag.2025.104795","DOIUrl":"10.1016/j.jag.2025.104795","url":null,"abstract":"<div><div>Cropland non-agriculturalization (CNA) monitoring is a typical change detection (CD) problem based on remote sensing imagery, aimed at tracking cropland outflow changes, which holds significant importance for cropland protection and food security. Recently, numerous advanced CD methods have been proposed to address the CNA problem. However, applying these methods to cross-regional or large-scale CNA detection presents several challenges: (1) Radiance-feature differences of croplands across regions i.e., crop type and phenology differences arising from variations in planting structures and seasonality; (2) Change-pattern differences of croplands across regions, i.e., differences in predominant change types resulting from distinct regional economic development characteristics. These cross-regional differences, when coupled together, result in insufficient adaptability of CD methods across regions. To address these issues, a Cross-Region Cropland Change Detection (CR-CLCD) framework with Multi-View Domain Adaptation (MVDA) is proposed. Specifically, Pattern Distribution Contrastive (PDC) sub-module achieves feature alignment from the semantic view by imposing contrastive constraints across inter-domain categories. Radiative Discrepancy Adversarial (RDA) sub-module, performs inter-domain global and local feature confusion by identifying regions of local uncertainty and applying enhanced adversarial training. MVDA is a flexible, plug-and-play domain adaptation module that can be seamlessly integrated with any existing change detection backbone network (e.g., CNN, Transformer), enabling rapid generalization to new data under unsupervised conditions. The experimental results demonstrate that the proposed CR-CLCD method achieves the best or second-best accuracy compared to other domain adaptation methods across different baselines.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"143 ","pages":"Article 104795"},"PeriodicalIF":8.6,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144858452","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":"A multi-tiered representativeness framework mitigating spatial scale effects in FVC validation: UAV-based assessment of global products in Qinghai-Tibetan Plateau alpine grasslands","authors":"Jianjun Chen ,&nbsp;Renjie Huang ,&nbsp;Lihui Luo ,&nbsp;Shuhua Yi ,&nbsp;Yu Qin ,&nbsp;Wenbo Qi ,&nbsp;Haotian You ,&nbsp;Xiaowen Han ,&nbsp;Guoqing Zhou","doi":"10.1016/j.jag.2025.104794","DOIUrl":"10.1016/j.jag.2025.104794","url":null,"abstract":"<div><div>Ground-measured fractional vegetation cover (FVC) data are critical for validating satellite-derived FVC products. However, spatial scale mismatches between ground plots and satellite pixels, compounded by the scarcity of field data in remote regions, introduce significant uncertainties in product validation. This study proposes a novel multi-tiered representativeness framework integrating three key indices: the absolute difference of spatial upscaling transformations, the heterogeneity of the surrounding environment, and FVC temporal stability. The framework’s implementation leveraged an unmanned aerial vehicle (UAV) observation network (870 monitoring plots, 2015–2024) within the alpine grassland ecosystem of the Three-River Source Region on the Qinghai-Tibet Plateau, classified into four levels, with levels 1–2 indicating high level and 3–4 lower. Our results reveal that both the environmental heterogeneity of monitoring plots and scale mismatches substantially impact the validation accuracy of FVC products. By applying the proposed framework, validation using high-level monitoring plots reduced uncertainty by approximately 40 % compared to using all monitoring plots (GEOV3: R² = 0.964, RMSE = 0.075 vs. R² = 0.830, RMSE = 0.138; GLASS: R² = 0.957, RMSE = 0.068 vs. R² = 0.812, RMSE = 0.121), highlighting its effectiveness in mitigating spatial representativeness errors. Furthermore, the validation results for two global FVC products (GEOV3 and GLASS) highlight systematic biases in their performance within alpine ecosystems. These findings advance validation methodologies for remote sensing products in heterogeneous landscapes and provide actionable insights for improving algorithm parameterization. The framework’s modular design enables adaptation to other critical validation scenarios requiring spatial representativeness quantification.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"143 ","pages":"Article 104794"},"PeriodicalIF":8.6,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144852130","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":"BayeSiamMTL: Uncertainty-aware multitask learning for post-disaster building damage assessment","authors":"Victor Hertel ,&nbsp;Omar Wani ,&nbsp;Christian Geiß ,&nbsp;Marc Wieland ,&nbsp;Hannes Taubenböck","doi":"10.1016/j.jag.2025.104759","DOIUrl":"10.1016/j.jag.2025.104759","url":null,"abstract":"<div><div>Accurate and timely building damage assessment (BDA) is critical for effective disaster response and recovery. However, existing machine learning approaches in this context do mostly not account for uncertainties, which are essential for ensuring trustworthy and transparent results. This study introduces a hybrid Bayesian deep learning framework with integrated uncertainty quantification to enhance BDA, thereby making model predictions more reliable and interpretable. We propose BayeSiamMTL, a novel Bayesian Siamese multitask learning architecture that combines deterministic segmentation of building footprints with probabilistic change detection for damage level classification. By encoding model parameters as probability distributions and utilizing variational inference with Monte Carlo approximation, BayeSiamMTL produces pixelwise posterior predictive distributions, providing detailed insights into both damage predictions and their associated uncertainties. Our analysis explores key aspects of Bayesian modeling and, to our knowledge, is the first to provide quantified insights into the model’s classification dynamics, revealing internal decision-making tendencies and sources of uncertainty. Additionally, we introduce confidence-informed damage maps in the form of stratified probabilities of damage clusters and minimum/maximum damage extents delineated from confidence intervals. Model performance is evaluated across multiple datasets to assess the impact of domain shifts and out-of-distribution samples. Experimental results show that BayeSiamMTL not only achieves a performance advantage over its deterministic counterpart but also exhibits significantly better generalization capabilities under domain shifts with a relative performance improvement of 42 %. While background pixels represent the primary source of confusion across all damage levels, our findings indicate that building destructions are more frequently confused with intact buildings rather than among varying degrees of damage.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"143 ","pages":"Article 104759"},"PeriodicalIF":8.6,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144852128","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":"High-precision airborne LiDAR remains essential for urban forestry: Revealing the limitations of recent large-scale canopy height products in urban contexts","authors":"Hesong Dong ,&nbsp;Zhibang Xu ,&nbsp;Jinzhou Wu ,&nbsp;Ting Lan ,&nbsp;Lin Wang ,&nbsp;Guofan Shao ,&nbsp;Lina Tang","doi":"10.1016/j.jag.2025.104791","DOIUrl":"10.1016/j.jag.2025.104791","url":null,"abstract":"<div><div>Urban trees play a crucial role in delivering ecosystem services, yet their structural complexity poses significant challenges for remote sensing. Although recent large-scale, open-access canopy height models (CHMs) offer potential alternatives to airborne LiDAR, their suitability for urban environments remains uncertain. This study systematically evaluated four prominent CHMs (Potapov, Lang, Tolan, and Malambo) using high-resolution airborne LiDAR data from Washington, D.C., with complementary analyses from four additional cities. We assessed their performance in both canopy classification and height prediction using classification and regression metrics, spatial autocorrelation, consistency tests, and explainable machine learning. Results revealed consistent limitations across products, including widespread misclassification of canopy, systematic tree height prediction biases — characterized by overestimation of low and underestimation of high canopies (the OLUH effect) — and pronounced spatial clustering of errors along urban–forest edges. Among the models, only the Lang CHM passed the Bland–Altman consistency test, showing marginal statistical agreement with reference data. Tree characteristic variables, especially canopy height itself, emerged as dominant drivers of height errors, while topography and built-up context also contributed. Consistent patterns observed across four additional cities indicated that these limitations are systemic rather than location-specific. We conclude that high-precision airborne LiDAR remains essential for urban forestry and recommend enhancing canopy height mapping techniques to better capture the structure of urban trees. A promising direction is the development of urban-specific CHMs with finer spatial and temporal resolution, improved temporal consistency, and integration with high-resolution imagery, contextual deep learning models, and local calibration strategies.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"143 ","pages":"Article 104791"},"PeriodicalIF":8.6,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144852129","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":"Accurate room layout estimation from multi-view panoramas with multi-label graph cut","authors":"Zhihua Hu ,&nbsp;Wanjie Lu ,&nbsp;Kao Zhang ,&nbsp;Helong Yang ,&nbsp;Yaoyang Wang ,&nbsp;Nannan Qin ,&nbsp;Yuxuan Liu ,&nbsp;Sisi Zlatanova","doi":"10.1016/j.jag.2025.104777","DOIUrl":"10.1016/j.jag.2025.104777","url":null,"abstract":"<div><div>Estimating room layout from panoramas is a new trend in the holistic reconstruction of the 3D environment. However, a single panorama is easily occluded by walls and furniture, making it hard to reconstruct the whole indoor room accurately and completely. Besides, deep learning room layout estimating methods often perform poorly in unseen scenes. To address this need, this paper proposes an accurate room layout estimation method from multi-view panoramas with multi-label graph cut. The proposed method takes full advantage of each panorama by utilizing multi-label graph cut. First, room layouts of each panorama are estimated with pre-trained deep-learning models and projected to the ground as the labels; then, a geometry-aware ray-casting method is utilized to obtain the initial floorplan; next, the initial floorplan is regularized by multi-label graph cut with the estimated labels from each panorama; in the end, the final layouts of each panorama is obtained by transforming the regularized floorplans and estimated ceiling heights into layouts with panorama geometry. Experiments in the recently released multi-view panoramas dataset show that the proposed method can regularize the initial floorplan to a floorplan with accurate geometry. Furthermore, the accuracy of the layouts surpassed the layout estimation accuracy of the single panorama deep learning models (HorizonNet and LGTNet) and the state-of-the-art self-training layout estimation models with multi-view panoramas by a large margin.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"143 ","pages":"Article 104777"},"PeriodicalIF":8.6,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144842708","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":"Prediction of the flood distribution caused by returning cropland to forest based on Generative Adversarial Network and multi-source remote sensing data","authors":"Nai Wei ,&nbsp;Yi Lin ,&nbsp;Hao Zheng","doi":"10.1016/j.jag.2025.104790","DOIUrl":"10.1016/j.jag.2025.104790","url":null,"abstract":"<div><div>Variations in Land Use and Land Cover (LULC) significantly influence flooding patterns, particularly through alterations in forest and cropland. However, existing flood prediction studies rarely utilize LULC maps as direct training inputs, typically generating probabilistic rather than spatially explicit flood distribution maps. This study addresses this gap by integrating Generative Adversarial Networks (GANs) with multi-source remote sensing data to predict flood distribution from LULC maps. We focus specifically on “Returning Cropland to Forest” initiative of China using 2020 remote sensing data from the Poyang Lake region. Flood impact ranges were extracted through Synthetic Aperture Radar (SAR) and optical data fusion, producing 3,972 LULC-flood map pairs for model training and validation. The multi-source fusion approach achieved substantial improvements in flood extraction accuracy. User accuracy increased from 60% (SAR-only) to 90% (SAR-optical fusion), while overall classification accuracy reached 89.76% with an F1 score of 94.58%. Model performance validation through Fréchet Inception Distance (FID) scores demonstrated high-quality flood map generation, with FID decreasing from 81.5 to 53 at epoch 45. Analysis revealed that reforestation ratios exceeding 50% significantly reduce flood occurrence. Optimal effectiveness was observed when forests are strategically positioned along rivers and lake edges rather than randomly distributed. These findings provide actionable guidance for planners to prioritize riparian zones for initial reforestation, then systematically achieve 50% reforestation ratios in flood-prone watersheds, integrating these targets within existing regulatory frameworks. This research contributes a robust framework for integrating LULC data into predictive flood models, advancing sustainable flood management and enhancing water resource security strategies.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"143 ","pages":"Article 104790"},"PeriodicalIF":8.6,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144829627","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":"Harmonized spatial-frequency domain synergy driven geospatial feature synthesis for enhanced SAR semantic segmentation","authors":"Minhong Sun ,&nbsp;Han Yang ,&nbsp;Zihan Xia ,&nbsp;Fengjiao Gan ,&nbsp;Zhao Huang ,&nbsp;Zhiwen Zheng ,&nbsp;Lou Zhao ,&nbsp;Chunshan Liu ,&nbsp;Zhaoyang Xu ,&nbsp;Yun Lin ,&nbsp;Guan Gui ,&nbsp;Xiaoshuai Zhang ,&nbsp;Xingru Huang ,&nbsp;Jin Liu","doi":"10.1016/j.jag.2025.104754","DOIUrl":"10.1016/j.jag.2025.104754","url":null,"abstract":"<div><div>The phase coherence of radar signals makes synthetic aperture radar (SAR) image analysis prone to significant challenges. Echo signal interference introduces speckle noise during imaging; noise appears as random fluctuations in pixel intensities. Besides, coherence exacerbates geometric distortions, complicating the accurate interpretation of spatial distributions within intricate geographic entities, thereby making it difficult to extract meaningful target information from SAR images. Addressing these challenges, this study introduces the DEcomposed-frequency PrOjection Network (Depo-Net), a segmentation-oriented model that mitigates SAR-specific interference through frequency-domain self-attention. It employs a dual-encoder structure for efficient semantic extraction and integrates Spatio-Frequency Synergistic Modulation (SFSM) to minimize speckle noise while maintaining structural integrity in the frequency domain. Additionally, the Harmonized Subspace Spectro-Temporal Attention (HSSTA) synthesizes Discrete Fourier and Wavelet Transform analyses to capture complex spatial correlations among geographic features. To mitigate noise amplification during decoding, the Pluri-frequency Mamba (purfMamba) module synergizes multi-dimensional spectral-spatial features, facilitating noise suppression during high-resolution restoration and maintaining a balance between global structure and local details. Results on three public SAR segmentation datasets demonstrate Depo-Net’s efficacy outperforming 22 previous State-of-the-Art (SOTA) methods while minimizing 95th Percentile Hausdorff Distance values. The complete code and model implementation is available on GitHub at <span><span>https://github.com/IMOP-lab/Depo-Net</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"143 ","pages":"Article 104754"},"PeriodicalIF":8.6,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144842707","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":"Value of microwave soil moisture and thermal-infrared evapotranspiration retrievals for the mapping of irrigation coverage","authors":"Wade T. Crow ,&nbsp;Martha C. Anderson ,&nbsp;John M. Volk ,&nbsp;Andreas Colliander","doi":"10.1016/j.jag.2025.104773","DOIUrl":"10.1016/j.jag.2025.104773","url":null,"abstract":"<div><div>To better monitor global water resources, and understand how they will vary in the future, it is critical to track the extent and intensity of irrigated agriculture. Within the past decade, there has been increased interest in the satellite-based detection of anomalous soil moisture (SM) and/or evapotranspiration (ET) signals associated with irrigation. However, little comparative information is available concerning the relative merits of available ET versus SM satellite products for this purpose. Such uncertainty has hampered the development of optimal monitoring strategies that appropriately integrate information acquired across a range of remote sensing resources. Here, using relatively more mature irrigation products derived from a combination of ground data and visible/near-infrared remote sensing as a reference, we compare the skill of microwave (MW) SM and thermal-infrared (TIR) ET satellite products for mapping coarse-scale (36-km) spatial variations in the proportion of land irrigated across the conterminous United States. Results suggest that, while spatial mapping skill exists in both products, the irrigation signal in satellite-based ET products is stronger, and easier to interpret, than the analogous signal in SM products. Since MW-based SM and TIR-based ET products appear to possess approximately equal spatial precision (i.e., spatial correlations with respect to true SM and ET, respectively), this difference is attributed to the stronger impact of irrigation on spatial ET patterns versus surface SM.</div><div><strong>Plain Language Summary:</strong> To better monitor global water resources, and understand how they will vary in the future, it is important to track the extent and intensity of irrigated agriculture. To do this, different satellite-based methods have been proposed to track irrigation from space. Unfortunately, there has been little comparison of these methods, and it is unclear which ones work best. Here, we compare methods for monitoring irrigation based on two different remote sensing techniques: thermal infrared and passive microwave. While information derived from both types of remote sensing is useful for irrigation monitoring, results show that thermal-infrared remote sensing contains more information. This insight will help improve future efforts to globally monitor irrigation using satellite-based sensors.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"143 ","pages":"Article 104773"},"PeriodicalIF":8.6,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144829626","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":"Shadow-robust unsupervised flood mapping via GMM-enhanced generalized dual-polarization flood index and topography features","authors":"Huifu Zhuang ,&nbsp;Zihao Tang ,&nbsp;Sen Du ,&nbsp;Peng Wang ,&nbsp;Hongdong Fan ,&nbsp;Ming Hao ,&nbsp;Zhixiang Tan","doi":"10.1016/j.jag.2025.104787","DOIUrl":"10.1016/j.jag.2025.104787","url":null,"abstract":"<div><div>Under cloudy and rainy conditions, Synthetic Aperture Radar (SAR) can provide essential data support for large-scale and high-timeliness flood monitoring. Although an optimal combination for Sentinel-1 VV-VH polarization data has been selected for flood mapping, its performance on complex terrains is not satisfactory, and it is unknown whether it can be further extended to generalize co- and cross-polarization data (not only VV-VH but also HH-HV). Therefore, we propose a shadow-robust unsupervised flood mapping method, called <strong>To</strong>pography and <strong>D</strong>ual-polarization <strong>Flo</strong>od <strong>M</strong>apping (TODFLOM). This method initially utilizes topography features to identify safe areas where floods are unlikely to occur in complex scenarios. Then, a Generalized Dual-Polarization Flood Index (GDPFI), based on the microwave scattering characteristics of complex inundation scenarios, is constructed to highlight flood features. Finally, GDPFI coupled with a Gaussian Mixture Model (GMM) is used for generating the flood map, enabling the method to effectively suppress mountain shadows in areas with slopes below 30°. The integration of topography features and GDPFI-GMM empowers TODFLOM to remove shadows impact with a 5° safety threshold. Experiments reveal that TODFLOM achieves an F1 score greater than 0.88 across all four flood datasets, outperforming other advanced methods in large-scale complex inundation scenarios.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"143 ","pages":"Article 104787"},"PeriodicalIF":8.6,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144829624","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}