ISPRS Journal of Photogrammetry and Remote Sensing最新文献

{"title":"Mamba-CNN hybrid Multi-scale ship detection Network driven by a Dual-perception feature of Doppler and Scattering","authors":"Gui Gao ,&nbsp;Caiyi Li ,&nbsp;Xi Zhang ,&nbsp;Bingxiu Yao ,&nbsp;Zhen Chen","doi":"10.1016/j.isprsjprs.2026.01.004","DOIUrl":"10.1016/j.isprsjprs.2026.01.004","url":null,"abstract":"<div><div>Ship detection is crucial for both military and civilian applications and is a key use of polarimetric SAR (PolSAR). While convolutional neural networks (CNNs) enhance PolSAR ship detection with powerful feature extraction, existing approaches still face challenges in discriminating targets from clutter, detecting multi-scale objects in complex scenes, and achieving real-time detection. To address these issues, we propose a Mamba-CNN hybrid Multi-scale ship detection Network driven by a Dual-perception feature of Doppler and Scattering. First, at the input feature level, a Dual-perception feature of Doppler and Scattering (DDS) is introduced, effectively differentiating ship and clutter pixels to enhance the network’s ship discrimination. Specifically, Doppler characteristics distinguish between moving and stationary targets, while scattering characteristics reveal fundamental differences between targets and clutter. Second, at the network architecture level, a Mamba-CNN hybrid Multi-scale ship detection Network (MCMN) is designed to improve multi-scale ship detection in complex scenarios. It uses a Multi-scale Information Perception Module (MIPM) to adaptively aggregate multi-scale features and a Local-Global Feature Enhancement Module (LGFEM) based on Mamba for long-range context modeling. MCMN remains efficient through feature grouping, pointwise and depthwise convolutions, meeting real-time requirements. Finally, extensive experiments on the GF-3 and SSDD datasets demonstrate the superiority of DDS and MCMN. DDS effectively distinguishes ships from clutter across scenarios. As an input feature, it boosts average F1-score and AP by 4.3% and 4.3%, respectively, over HV intensity, and outperforms other polarization features. MCMN achieves state-of-the-art results, improving AP by 1.2% and 0.8% on the two datasets while reducing parameters by 1.29M, FLOPs by 1.5G, and inference time by 59.2%.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"232 ","pages":"Pages 656-674"},"PeriodicalIF":12.2,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925541","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mapping melliferous tree species in Kenya via one-class classification with hyperspectral unsupervised domain adaptation","authors":"Zhaozhi Luo ,&nbsp;Janne Heiskanen ,&nbsp;Ilja Vuorinne ,&nbsp;Ian Ocholla ,&nbsp;Shiqi Zhang ,&nbsp;Saana Järvinen ,&nbsp;Xinyu Wang ,&nbsp;Yanfei Zhong ,&nbsp;Petri Pellikka","doi":"10.1016/j.isprsjprs.2025.12.028","DOIUrl":"10.1016/j.isprsjprs.2025.12.028","url":null,"abstract":"<div><div>The beekeeping sector holds significant potential for livelihood diversification among the agropastoral communities in Kenya. Melliferous tree species play a critical role by providing essential nectar sources for bees. However, limited knowledge of their precise spatial distributions constrains the full development of beekeeping. One-class classification (OCC) offers a practical solution for detecting single target species without requiring extensive labeled data from other classes. Although existing OCC methods perform well in trained domains, the generalization capability to unseen domains remains limited due to domain shift. To address these challenges, this study proposes a hyperspectral unsupervised domain adaptation OCC framework (HyUDA-One) for tree species mapping using airborne hyperspectral imagery and laser scanning data. The spatial–spectral regularized pseudo-positive learning was designed to mitigate domain shift and improve model generalizability. The effectiveness of HyUDA-One was demonstrated by mapping three key melliferous tree species in two savanna landscapes in southern Kenya. The results show that HyUDA-One significantly improves performance in unlabeled domains. The F1-scores of 0.788, 0.845, and 0.768 were achieved for <em>Senegalia mellifera</em>, <em>Vachellia tortilis</em>, and <em>Commiphora africana</em> in the trained domain, respectively. In the untrained domain, the F1-scores of <em>Senegalia mellifera</em> and <em>Vachellia tortilis</em> were 0.756 and 0.884, respectively. The distribution maps revealed the spatial patterns of these melliferous tree species and the nectar source availability, offering an important reference for sustainable beekeeping development in savanna landscapes. Furthermore, the proposed framework can potentially be extended to other mapping applications, such as invasive species detection.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"232 ","pages":"Pages 638-655"},"PeriodicalIF":12.2,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A spectral index using generic global endmembers from Landsat multispectral data for mapping urban areas","authors":"Ruiyi Zhao ,&nbsp;Cai Cai ,&nbsp;Xinfan Cai ,&nbsp;Peijun Li","doi":"10.1016/j.isprsjprs.2025.12.025","DOIUrl":"10.1016/j.isprsjprs.2025.12.025","url":null,"abstract":"<div><div>Quantifying urban land cover from space is crucial for studying and understanding its spatial distribution and changes, as well as for assessing the impact of these changes on environmental and socio-economic dynamics worldwide. Owing to the diversity and spectral heterogeneity of urban reflectance, spectrally mixed pixels predominate in moderate resolution multispectral images. The standardized linear spectral mixture model from Landsat multispectral images, which represents the radiance measurements of mixed pixels as linear mixtures of generic global Substrate (S), Vegetation (V), and Dark Surfaces (D) endmember radiances, offers an effective method for characterizing urban reflectance. Based on the analysis of SVD endmember fractions of urban land and other land cover types, this study proposes a spectral index using Landsat global SVD endmembers, termed the Urban Index using Global Endmembers (GEUI), to highlight and map urban land. GEUI is evaluated through comparisons with five established spectral indices: including Normalized Difference Built-up Index (NDBI), Index-based Built-up Index (IBI), Biophysical Composition Index (BCI), Built-up Land Features Extraction Index (BLFEI), and Urban Composition Index (UCI), all of which rely on pure spectral signatures of urban pixels. Additionally, GEUI is compared to two deep learning methods in urban area mapping, i.e., two-dimensional convolutional neural network (2D CNN) and one-dimensional CNN (1D CNN). The results demonstrate that the proposed GEUI outperforms these comparative indices in qualitative evaluation, separability analysis, and urban land mapping, and also showed superior performance in urban land mapping compared to CNN methods. GEUI achieved overall accuracies ranging from 84.36% to 93.02% and F-scores between 84.80% and 92.64%, obtaining the highest accuracy in half of the study urban areas. Since S, V, and D endmembers used in GEUI are globally available, the proposed GEUI has the advantage of being applicable across diverse locations and times. Furthermore, GEUI can be readily extended to other broadband multispectral data, such as Sentinel-2 and MODIS. Therefore, the proposed GEUI provides an effective variable for mapping urban land and holds the potential<!--> <!-->for diverse<!--> <!-->urban applications.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"232 ","pages":"Pages 613-637"},"PeriodicalIF":12.2,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925471","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Progressive uncertainty-guided network for binary segmentation in high-resolution remote sensing imagery","authors":"Jiepan Li ,&nbsp;Wei He ,&nbsp;Ting Hu ,&nbsp;Minghao Tang ,&nbsp;Liangpei Zhang","doi":"10.1016/j.isprsjprs.2026.01.010","DOIUrl":"10.1016/j.isprsjprs.2026.01.010","url":null,"abstract":"<div><div>Binary semantic segmentation in remote sensing (RS) imagery faces persistent challenges due to complex object appearances, ambiguous boundaries, and high similarity between foreground and background, all of which introduce significant uncertainty into the prediction process. Existing approaches often treat uncertainty as either a global attribute or a pixel-level estimate, overlooking the critical role of spatial and contextual interactions. To address these limitations, we propose the <strong>Progressive Uncertainty-Guided Segmentation Network (PUGNet)</strong>, a unified framework that explicitly models uncertainty in a context-aware manner. PUGNet decomposes uncertainty into three distinct components: <strong>foreground uncertainty</strong>, <strong>background uncertainty</strong>, and <strong>contextual uncertainty</strong>. This tripartite modeling enables more precise handling of local ambiguities and global inconsistencies. We adopt a coarse-to-fine decoding strategy that progressively refines features through two specialized modules. The <strong>Dynamic Uncertainty-Aware Module</strong> enhances regions of high foreground and background uncertainty using Gaussian-based modeling and contrastive learning. The <strong>Entropy-Driven Refinement Module</strong> quantifies contextual uncertainty via entropy and facilitates adaptive refinement through multi-scale context aggregation. Extensive experiments on ten public benchmark datasets, covering both single-temporal (<em>e.g.</em>, building and cropland extraction) and bi-temporal (<em>e.g.</em>, building change detection) binary segmentation tasks, demonstrate that PUGNet consistently achieves superior segmentation accuracy and uncertainty reduction, establishing a new state of the art in RS binary segmentation. The full implementation of the proposed framework and all experimental results can be accessed at <span><span>https://github.com/Henryjiepanli/PU_RS</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"232 ","pages":"Pages 561-577"},"PeriodicalIF":12.2,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925511","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An enhanced spatiotemporal prediction method on landslide displacement with LDP-ConvFormer and MT-InSAR observations","authors":"Jianao Cai ,&nbsp;Dongping Ming ,&nbsp;Feng Liu ,&nbsp;Wenyi Zhao ,&nbsp;Mingzhi Zhang ,&nbsp;Xiao Ling ,&nbsp;Mengyuan Zhu ,&nbsp;Lu Xu ,&nbsp;Tingting Lu ,&nbsp;Ningjie Liu ,&nbsp;Yanfei Wei ,&nbsp;Ming Huang","doi":"10.1016/j.isprsjprs.2025.12.027","DOIUrl":"10.1016/j.isprsjprs.2025.12.027","url":null,"abstract":"<div><div>Landslide Displacement Prediction (LDP) implementation for Landslide Early Warning Systems (LEWS) using the Multi-Temporal Interferometric Synthetic Aperture Radar (MT-InSAR) technique poses significant challenges in the Three Gorges Reservoir Area (TGRA). On the one hand, the limited revisit frequency of satellites fails to satisfy the high-frequency monitoring requirements of LEWS. On the other hand, traditional LDP methods concentrate on single-point modeling. It neglects the spatial correlation between displacement points and the landslide surface. To enhance the low-frequency MT-InSAR observations, this paper proposes a new hybrid algorithm that integrates the Kalman Filter (KF) and LDP-ConvFormer to achieve enhanced spatiotemporal LDP. First, multi-orbit MT-InSAR measurements are transformed to downslope displacement. Subsequently, the multi-orbit downslope displacements are integrated by KF to generate time series data with enhanced temporal resolution (5/7-day intervals). The KF estimations indicate that the integrated higher-resolution time series achieves high accuracy, with an RMSE of 0.431 cm and an R² of 0.974 compared to GNSS. Finally, to overcome the limitation of single-point modeling, a novel LDP-ConvFormer is constructed for enhanced spatiotemporal LDP. The Spatiotemporal Displacement Prediction Transformer (STDP-Former) employs Local Spatial Multi-Head Self-Attention (LSMHSA) and Temporal Multi-Head Self-Attention (TMHSA) to capture the displacement dependencies between different spatial locations at the same time steps and temporal relationships across different time steps, respectively. Additionally, the spatiotemporal feature map is decomposed into trend and periodic components, which are modeled separately and then summed for final predictions. Experimental results demonstrate that the constructed model can accurately establish the nonlinear relationship between the landslide displacement and its triggering factors. The LDP-ConvFormer outperforms benchmark methods, achieving RMSE: 46.29 mm, MAE: 26.7 mm, SSIM: 0.8187, PSNR: 35.62, R²: 0.9574, and EVar: 0.9603. Moreover, LDP-ConvFormer shows notable superiority in LDP over medium to long periods (60-90d) in the TGRA. The enhanced spatiotemporal LDP method provides extremely valuable reference for LEWS of translational landslides in the TGRA.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"232 ","pages":"Pages 594-612"},"PeriodicalIF":12.2,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925543","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mapping land uses following tropical deforestation with location-aware deep learning","authors":"Jan Pišl ,&nbsp;Gencer Sumbul ,&nbsp;Gaston Lenczner ,&nbsp;Camilo Zamora ,&nbsp;Martin Herold ,&nbsp;Jan Dirk Wegner ,&nbsp;Devis Tuia","doi":"10.1016/j.isprsjprs.2025.12.007","DOIUrl":"10.1016/j.isprsjprs.2025.12.007","url":null,"abstract":"<div><div>The rates of tropical deforestation remain alarmingly high. To enable effective, targeted policy responses, detailed data on its driving forces is needed—each deforestation event needs to be attributed to an agricultural commodity or another land use. Remote sensing allows us to monitor land use conversion following deforestation, providing a proxy of drivers. However, recognizing individual commodities is challenging due to spectral similarities, the limited spatial resolution of free satellite imagery, and limited labeled data. To tackle these challenges, we propose a deep learning, multi-modal approach for the recognition of post-deforestation land uses from a time series of Sentinel-2 images, geographic coordinates, and country-level statistics of deforestation drivers. To integrate the modalities, we design a Transformer-based model with modality-specific encoders. The approach reaches 87% accuracy, an improvement of 10% over the image-only baseline, with little increase in data volume, computations, and model size. It works well in low-data regimes, and can be easily extended to include other modalities. Overall, this work contributes towards detailed, repeatable, and scalable mapping of deforestation landscapes, providing necessary data for the design and implementation of targeted interventions to protect tropical forests.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"232 ","pages":"Pages 578-593"},"PeriodicalIF":12.2,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925544","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Beyond synthetic scenarios: Weakly-supervised super-resolution for spatiotemporally misaligned remote sensing images","authors":"Quanyi Guo ,&nbsp;Rui Liu ,&nbsp;Yangtian Fang ,&nbsp;Yi Gao ,&nbsp;Jun Chen ,&nbsp;Xin Tian","doi":"10.1016/j.isprsjprs.2025.12.019","DOIUrl":"10.1016/j.isprsjprs.2025.12.019","url":null,"abstract":"<div><div>Deep learning-based remote sensing image super-resolution is crucial for enhancing the spatial resolution of Earth observation data. Due to the absence of perfectly aligned pairs of high- and low-resolution remote sensing images, most existing supervised and self-supervised approaches rely on synthetic degradation models or internal structural consistency to generate training data. Consequently, these methods suffer from the domain gap between synthetic and real datasets, which limits their ability to model realistic degradation and degrades their performance in real scenes. To overcome this challenge, we propose STANet, a weakly-supervised super-resolution method for spatiotemporally misaligned remote sensing images. In particular, STANet directly utilizes images of the same region captured by multiple satellites at different resolutions as datasets, to boost the real remote sensing image super-resolution performance. However, this approach also introduces new challenges related to spatiotemporal misalignment. To address this, we design a spatiotemporal align module that includes a Scale Align Module (SAM) and a Temporal Align Module (TAM). SAM uses affine transformations to align spatial features at both the pixel and global levels, while TAM applies window-based attention to adjust the weight of image content, mitigating the misleading effects of temporal misalignment on results. Besides, we also design a style encoder based on contrastive learning and a structure encoder based on variational inference, which guide SAM and TAM for feature alignment and enhance adaptability. Finally, the feature-aligned output, after upsampling, are fused with the high-frequency-enhancing output of the texture transfer module through the weighted fusion module to generate the super-resolution image. Extensive experiments on synthetic datasets based on AID and RSSR25, real datasets captured by GaoFen satellites, and cross-satellite experiments on Landsat-8 datasets demonstrate STANet’s superiority over other state-of-the-art methods.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"232 ","pages":"Pages 524-541"},"PeriodicalIF":12.2,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145902921","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advancing table beet root yield estimation via unmanned aerial systems (UAS) multi-modal sensing","authors":"Mohammad S. Saif ,&nbsp;Robert Chancia ,&nbsp;Sean P. Murphy ,&nbsp;Sarah Pethybridge ,&nbsp;Jan van Aardt","doi":"10.1016/j.isprsjprs.2025.12.026","DOIUrl":"10.1016/j.isprsjprs.2025.12.026","url":null,"abstract":"<div><div>Unmanned aerial systems (UAS) offer significant potential to improve agricultural practices due to their multi-modal payload capacity, ease of deployment, and lower cost. However, there is a need to expand UAS capabilities by including root crops, offering robust, growth-stage-independent models, and providing a comprehensive assessment of various imaging systems, i.e., identifying application-specific sensing modalities. This study aims to tackle those challenges by presenting a unified Gaussian Process Regression (GPR) model for predicting end-of-season table beet (a subterranean root crop) yield using UAS-derived spectral and structural features, combined with meteorological data, while remaining robust to flight and harvest timing. Field trials were conducted at Cornell AgriTech in Geneva, NY during the 2021 and 2022 growing seasons. UAS flights captured five-band (475, 560, 668, 717, and 840 nm) multispectral imagery, hyperspectral imagery (400–1000 nm), and light detection and ranging (LiDAR) data at multiple times throughout the season. Our model achieved <em>R²_test</em> = 0.81 and MAPE_test = 15.7 % using only multispectral imagery, while the hyperspectral + LiDAR model attained <em>R²_test</em> = 0.79 and MAPE_test = 17.4 %, which is comparable to recent root yield modeling studies using UAS data. Shapely analysis was performed to gain further insight into model behavior. This analysis revealed canopy volume information to contain high relative importance, as compared to other features, for table beet root yield estimation. Our study demonstrated that UAS-based imaging, combined with a unified machine learning model, can effectively predict root crop yield, providing a scalable and transferable approach for precision agriculture.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"232 ","pages":"Pages 542-560"},"PeriodicalIF":12.2,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MARSNet: A Mamba-driven adaptive framework for robust multisource remote sensing image matching in noisy environments","authors":"Weipeng Jing ,&nbsp;Peilun Kang ,&nbsp;Donglin Di ,&nbsp;Jian Wang ,&nbsp;Yang Song ,&nbsp;Chao Li ,&nbsp;Lei Fan","doi":"10.1016/j.isprsjprs.2025.12.021","DOIUrl":"10.1016/j.isprsjprs.2025.12.021","url":null,"abstract":"<div><div>Semi-dense matching of multi-source remote sensing images under noise interference remains a challenging task. Existing detector-free methods often exhibit low efficiency and reduced performance when faced with large viewpoint variations and significant noise disturbances. Due to the inherent noise and modality differences in multi-source remote sensing images, the accuracy and robustness of feature matching are substantially compromised. To address this issue, we propose a hybrid network for multi-source remote sensing image matching based on an efficient and robust Mamba framework, named MARSNet. The network achieves efficient and robust matching through the following innovative designs: First, it leverages the efficient Mamba network to capture long-range dependencies within image sequences, enhancing the modeling capability for complex scenes. Second, a frozen pre-trained DINOv2 foundation model is introduced as a robust feature extractor, effectively improving the model’s noise resistance. Finally, an adaptive fusion strategy is employed to integrate features, and the Mamba-like linear attention mechanism is adopted to refine the Transformer-based linear attention, further enhancing the efficiency and expressive power for long-sequence processing. To validate the effectiveness of the proposed method, extensive experiments were conducted on multi-source remote sensing image datasets, covering various scenarios such as noise-free, additive random noise, and periodic stripe noise. The experimental results demonstrate that the proposed method achieves significant improvements in matching accuracy and robustness compared to state-of-the-art methods. Additionally, by performing pose error evaluation on a large-scale general dataset, the superior performance of the proposed method in 3D reconstruction is validated, complementing the test results from the multi-source remote sensing dataset, thereby providing a more comprehensive assessment of the method’s generalization ability and robustness.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"232 ","pages":"Pages 457-475"},"PeriodicalIF":12.2,"publicationDate":"2026-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145903503","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Spatially Masked Adaptive Gated Network for multimodal post-flood water extent mapping using SAR and incomplete multispectral data","authors":"Hyunho Lee,&nbsp;Wenwen Li","doi":"10.1016/j.isprsjprs.2025.12.023","DOIUrl":"10.1016/j.isprsjprs.2025.12.023","url":null,"abstract":"<div><div>Mapping water extent during a flood event is essential for effective disaster management throughout all phases: mitigation, preparedness, response, and recovery. In particular, during the response stage, when timely and accurate information is important, Synthetic Aperture Radar (SAR) data are primarily employed to produce water extent maps. This is because SAR sensors can observe through cloud cover and operate both day and night, whereas Multispectral Imaging (MSI) data, despite providing higher mapping accuracy, are only available under cloud-free and daytime conditions. Recently, leveraging the complementary characteristics of SAR and MSI data through a multimodal approach has emerged as a promising strategy for advancing water extent mapping using deep learning models. This approach is particularly beneficial when timely post-flood observations, acquired during or shortly after the flood peak, are limited, as it enables the use of all available imagery for more accurate post-flood water extent mapping. However, the adaptive integration of partially available MSI data into the SAR-based post-flood water extent mapping process remains underexplored. To bridge this research gap, we propose the Spatially Masked Adaptive Gated Network (SMAGNet), a multimodal deep learning model that utilizes SAR data as the primary input for post-flood water extent mapping and integrates complementary MSI data through feature fusion. In experiments on the C2S-MS Floods dataset, SMAGNet consistently outperformed other multimodal deep learning models in prediction performance across varying levels of MSI data availability. Specifically, SMAGNet achieved the highest IoU score of 86.47% using SAR and MSI data and maintained the highest performance with an IoU score of 79.53% even when MSI data were entirely missing. Furthermore, we found that even when MSI data were completely missing, the performance of SMAGNet remained statistically comparable to that of a U-Net model trained solely on SAR data. These findings indicate that SMAGNet enhances the model robustness to missing data as well as the applicability of multimodal deep learning in real-world flood management scenarios. The source code is available at <span><span>https://github.com/ASUcicilab/SMAGNet</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"232 ","pages":"Pages 492-508"},"PeriodicalIF":12.2,"publicationDate":"2026-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145902910","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}