IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing最新文献

{"title":"Toward Integrating ChatGPT Into Satellite Image Annotation Workflows: A Comparison of Label Quality and Costs of Human and Automated Annotators","authors":"Jacob Beck;Lukas Malte Kemeter;Konrad Dürrbeck;Mohamed Hesham Ibrahim Abdalla;Frauke Kreuter","doi":"10.1109/JSTARS.2025.3528192","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3528192","url":null,"abstract":"High-quality annotations are a critical success factor for machine learning (ML) applications. To achieve this, we have traditionally relied on human annotators, navigating the challenges of limited budgets and the varying task-specific expertise, costs, and availability. Since the emergence of large language models (LLMs), their popularity for generating automated annotations has grown, extending possibilities and complexity of designing an efficient annotation strategy. Increasingly, computer vision capabilities have been integrated into general-purpose LLMs like ChatGPT. This raises the question of how effectively LLMs can be used in satellite image annotation tasks and how they compare to traditional annotator types. This study presents a comprehensive investigation and comparison of various human and automated annotators for image classification. We evaluate the feasibility and economic competitiveness of using the ChatGPT4-V model for a complex land usage annotation task and compare it with alternative human annotators. A set of satellite images is annotated by a domain expert and 15 additional human and automated annotators, differing in expertise and costs. Our analyzes examine the annotation quality loss between the expert and other annotators. This comparison is conducted through, first, descriptive analyzes, second, fitting linear probability models, and third, comparing F1-scores. Ultimately, we simulate annotation strategies where samples are split according to an automatically assigned certainty score. Routing low-certainty images to human annotators can cut total annotation costs by over 50% with minimal impact on label quality. We discuss implications regarding the economic competitiveness of annotation strategies, prompt engineering, and the task-specificity of expertise.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"4366-4381"},"PeriodicalIF":4.7,"publicationDate":"2025-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10841407","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143105438","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advanced Machine Learning Ensembles for Improved Precipitation Forecasting: The Modified Stacking Ensemble Strategy in China","authors":"Tiantian Tang;Yifan Wu;Yujie Li;Lexi Xu;Xinyi Shi;Haitao Zhao;Guan Gui","doi":"10.1109/JSTARS.2025.3528475","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3528475","url":null,"abstract":"Accurate and reliable precipitation forecasting is vital for effective water resource management and disaster mitigation, especially in geographically diverse and climatically complex regions like China. This study proposes an advanced methodology for medium- and long-term hydrological forecasting by integrating multiple machine learning models through a modified stacking ensemble strategy (MSES). We developed and compared five deterministic precipitation forecasting models, including elastic net regression (ENR), support vector regression, random forest, extreme gradient boosting, and light gradient boosting to provide forecasts with lead times ranging from 0 to 5 months at a spatial resolution of 0.5<inline-formula><tex-math>$^circ$</tex-math></inline-formula>. The MSES was then evaluated against the traditional Bayesian model averaging (BMA) approach. Our comprehensive evaluation, based on deterministic forecasting metrics such as the anomaly correlation coefficient (ACC), mean squared skill score (MSSS), and Graded Precipitation Score (Pg), demonstrated the MSES outperformed individual models and the BMA method. The MSES achieved ACC scores between 0.6 and 0.9 for lead time (LT) <inline-formula><tex-math>$= 0$</tex-math></inline-formula> month, with an average of around 0.8 for LT <inline-formula><tex-math>$= 2$</tex-math></inline-formula> months. The MSSS for MSES was above 0.5 in more than half of the evaluations, and the Pg score was consistently above 80, indicating high accuracy in precipitation magnitude prediction. These findings highlight the promise of advanced machine learning strategies like MSES in improving the accuracy and robustness of precipitation forecasts, addressing critical needs in water resource management and disaster mitigation in China.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"4242-4254"},"PeriodicalIF":4.7,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10839126","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143105448","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Estimation of Forest Leaf Area Index Based on GEE Data Fusion Method","authors":"Xinyi Liu;Li He;Zhengwei He;Yun Wei","doi":"10.1109/JSTARS.2025.3528429","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3528429","url":null,"abstract":"Implementing forest protection measures, such as afforestation, can be an effective approach toward slowing down the increase of CO<inline-formula><tex-math>$_{2}$</tex-math></inline-formula> concentration and attaining carbon neutrality. The estimation of forest parameters is of great significance in understanding regional and global climate change patterns, and the Forest Leaf Area Index (LAI) is a crucial parameter. Current LAI products are mostly generated by moderate-resolution remote sensing data which does not meet the precision requirements for mountain forest ecosystems. To overcome this issue, there is an urgent need for higher resolution LAI data. This article proposes a data fusion method to map LAI in Wolong Nature Reserve that utilizes Sentinel-2 reflectance data, solar sensor geometry parameters, and vegetation indices extracted from the Google Earth Engine platform, along with canopy height data derived from canopy height estimation models in previous studies, combined with GLASS LAI V6 to estimate LAI using the random forest algorithm. The resulting LAI distribution map was plotted at a resolution of 20 m. The study demonstrated that incorporating canopy heights into the estimation model led to an <inline-formula><tex-math>$R^{2}$</tex-math></inline-formula> model accuracy of greater than 0.83. The 20-m resolution LAI map increased spatial details compared to the moderate-resolution LAI map, making it more suitable for mountain forest ecosystems that exhibit significant spatial heterogeneity.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"4510-4524"},"PeriodicalIF":4.7,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10839140","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143105956","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Remote Sensing of Sun-Induced Fluorescence in a Deep Lake: Disentangling Quenching Mechanisms Improves Relationship With Chlorophyll-a Concentration Estimates","authors":"Remika S. Gupana;Daniel Odermatt;Abolfazl Irani Rahaghi;Camille Minaudo;Mortimer Werther;Claudia Giardino;Alexander Damm","doi":"10.1109/JSTARS.2025.3528911","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3528911","url":null,"abstract":"Sun-induced fluorescence (SIF) from phytoplankton has historically been used as a proxy for chlorophyll-a (chl-a) concentration estimates in water bodies using optical earth observation data. However, the relationship is often affected by spectral features caused by elastic scattering, and by the shifting incidence of different fluorescence quenching mechanisms. This study found that disentangling photochemical quenching (PQ) and nonphotochemical quenching (NPQ) cases improves SIF-based chl-a estimates. Furthermore, we defined strategies that can distinguish the two quenching mechanisms. We assembled a unique dataset collected between 2018 and 2022 by an autonomous profiler in Lake Geneva (Western Europe). We used NPQ-influenced chl-a estimates from the fluorometer and NPQ-corrected chl-a estimates to distinguish between PQ and NPQ cases. The correlation between SIF yield and chl-a is weak when considering the entire dataset (<italic>R</i>² = 0.37 and median absolute percentage difference (MAPD) = 74%). It increases strongly when comparing PQ (<italic>R</i>² = 0.72 and MAPD = 49%) and NPQ cases (<italic>R</i>² = 0.48 and MAPD = 68%) separately. Analyzing a subset of in situ measurements acquired around Sentinel-3 overpasses (±3 h) improved the performance metrics for both PQ (<italic>R</i>² = 0.82 and MAPD = 35%) and NPQ cases (<italic>R</i>² = 0.43 and MAPD = 61%). However, when applying the same approach to Sentinel-3 Ocean and Land Color Instrument data, we found that the errors in remote sensing reflectance products disable such an adaptation. We conclude that enhanced atmospheric correction in the red-to-near-infrared region for oligo-mesotrophic lakes is needed to demonstrate the upscaling of our in-situ-based results. This will enhance satellite-based SIF yield retrievals and, subsequently, obtain SIF-related phytoplankton physiology products.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"4410-4426"},"PeriodicalIF":4.7,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10839553","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143106183","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Nonlinear Hybrid Algorithm for Retrieving Land Surface Temperatures From Chinese Atmospheric Environment Monitoring Satellite Thermal Infrared Data","authors":"Yichao Li;Hang Zhao;Kun Li;Jian Zeng;Qiongqiong Lan;Qijin Han;You Wu;Yonggang Qian","doi":"10.1109/JSTARS.2025.3528517","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3528517","url":null,"abstract":"Land surface temperature (LST) is a crucial parameter for representing the earth's surface energy balance. Thermal infrared remote sensing is the primary method for rapidly retrieving LST over large areas. The Chinese Atmospheric Environment Monitoring Satellite (DQ-1) is equipped with the wide swath imager (WSI), which includes three thermal infrared bands capable of providing global LST retrieval. This article introduces a nonlinear hybrid algorithm that combines the split-window (SW) algorithm and the temperature and emissivity separation (TES) algorithm, and the accuracies of the three algorithms, including hybrid, SW and TES algorithm are analyzed. The results demonstrated that the root mean square errors of LST for SW, TES, and hybrid algorithm are approximately 2.11, 1.78, and 1.64 K, with mean absolute errors (of 1.72, 1.40, and 1.21 K using in situ measurements from the SURFRAD sites. Cross-validation with moderate-resolution imaging spectroradiometer (MODIS) LST products showed that the hybrid algorithm outperforms the SW and TES algorithms in retrieving LST, achieving reductions in LST error of 0.43 and 0.16 K at the Qinghai Lake site, and 0.67 and 0.06 K at the Dunhuang site, respectively. In summary, this study demonstrates that the nonlinear hybrid algorithm can accurately estimate LST from DQ1/WSI data.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"4050-4059"},"PeriodicalIF":4.7,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10839136","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143361109","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SFMHANet: Surface Fitting Constrained Multidimensional Hybrid Attention Network for Aero-Optics Thermal Radiation Effect Correction","authors":"Yu Shi;ShanLin Niu;Lei Wang;Liang Ye;YaoZong Zhang;HanYu Hong","doi":"10.1109/JSTARS.2025.3528630","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3528630","url":null,"abstract":"When an aircraft is flying at hypervelocity in the atmosphere, the airflow and the optical cowl rub against each other, and the airflow's kinetic energy in the boundary layer is transformed into thermal energy, which causes the cowl's surface temperature to rise nonuniformly and produces thermal radiation interference with the imaging system of the detector. In practical application scenarios, the aero-optical thermal radiation patterns in degraded images are not fixed, and types of aero-optics thermal radiation are more variable and complex. In order to handle multiple types of aero-optics thermal radiation effects effectively and to combine the advantages of image prior constraints and deep learning networks, we propose a surface fitting constrained multidimensional hybrid attention aero-optics thermal radiation correction network (SFMHANet) in this article. First, according to the characteristics of the aero-optics thermal radiation bias field belonging to low frequency, we initially estimate the aero-optics thermal radiation bias field using biharmonic spline interpolation surface fitting based on wavelet decomposition. Second, we design a multidimensional hybrid attention aero-optics thermal radiation correction network constrained by the supervision of aero-optics thermal radiation bias field for asymmetric information exchange. Finally, to achieve cross-dimensional information interaction of features, we propose a multidimensional hybrid attention module, a second-order pooling channel attention block, and a cross-convolution spatial attention block in the correction network. According to experiments on aero-optics thermal radiation correction of simulated and real degraded images, the SFMHANet can correct the aero-optics thermal radiation effects of multitype degraded images in comparison to other existing methods.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"4569-4584"},"PeriodicalIF":4.7,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10839098","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143106045","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Nonlinear Waveform Sensing for Cognitive Radar Based on Reinforcement Learning","authors":"Peikun Zhu;Xu Si;Jiachen Han;Jing Liang","doi":"10.1109/JSTARS.2025.3528659","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3528659","url":null,"abstract":"Cognitive radar automatically adjusts its waveform via ceaseless interaction with the environment and learning from the experience. Compared with the linear frequency modulation (LFM) that has been commonly adopted in cognitive radars, the nonlinear FM (NLFM) signal has more flexible frequency variation and small time delay–Doppler coupling. In this work, we propose an NLFM cognitive radar based on reinforcement learning for target sensing. Specifically, a radar waveform selection framework is proposed via the interactive multimodel. It embraces the Riccati equation and Riccati-like iterative calculations to obtain the prediction error covariance (PEC) and the prediction Bayesian Cramér–Rao lower bound (PBCRLB), respectively, which are used to guide the optimal waveform design. With PEC or PBCRLB, an entropy reward Q-learning method is also proposed for joint waveform parameter selection (JWPS) and pure waveform parameter selection from the NLFM library. Simulations show that both the time complexity and tracking accuracy of PEC-based Q-learning JWPS outperform that of the PBCRLB method. Furthermore, PXIe-5785 is utilized to construct a cognitive radar platform and conduct field experiments for nonlinear waveform sensing, which confirms that nonlinear waveforms are more effective than linear waveforms in target localization.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"4821-4835"},"PeriodicalIF":4.7,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10838705","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143361108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing Remote Sensing Semantic Segmentation Accuracy and Efficiency Through Transformer and Knowledge Distillation","authors":"Kang Zheng;Yu Chen;Jingrong Wang;Zhifei Liu;Shuai Bao;Jiao Zhan;Nan Shen","doi":"10.1109/JSTARS.2025.3525634","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3525634","url":null,"abstract":"In semantic segmentation tasks, the transition from convolutional neural networks (CNNs) to transformers is driven by the latter's superior ability to capture global semantic information in remote sensing images. However, most transformer methods face challenges such as slow inference speed and limitations in capturing local features. To address these issues, this study designs a hybrid approach that integrates knowledge distillation with a combination of CNN and transformer to enhance semantic segmentation in remote sensing images. First, this article proposes the dual-path convolutional transformer network (DP-CTNet) with a dual-path structure to leverage the strengths of both CNN and transformers. It incorporates a feature refinement module to optimize the transformer's feature learning, and a feature fusion module to effectively merge CNN and transformer features, preventing the insufficient learning of local features by the transformer. Then, DP-CTNet serves as the teacher model, and pruning and knowledge distillation are employed to create efficient DP-CTNet (EDP-CTNet) with superior segmentation speed and accuracy. Angle knowledge distillation (AKD) is proposed to enhance the feature migration learning of DP-CTNet during knowledge distillation, leading to improved EDP-CTNet performance. Experimental results demonstrate that DP-CTNet thoroughly combines the respective advantages of CNN and Transformer, maintaining local detail features while learning extensive sequential semantic information. EDP-CTNet not only delivers impressive segmentation speed but also exhibits excellent segmentation accuracy following AKD training. In comparison to other models, the two models proposed in this article notably distinguish themselves in terms of accuracy and result visualization.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"4074-4092"},"PeriodicalIF":4.7,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10839278","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143106182","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Geodetic Evidence of the Interannual Fluctuations and Long-Term Trends Over the Antarctic Ice Sheet Mass Change","authors":"Yuanjin Pan;Xiaohong Zhang;Jiashuang Jiao;Hao Ding;C. K. Shum","doi":"10.1109/JSTARS.2025.3528516","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3528516","url":null,"abstract":"The spatiotemporal characteristics of the Antarctic ice sheet (AIS), as constrained by geodetic observations, provide us with a deeper understanding of the current evolution of ice mass balance. However, it still needs further in-depth research on interannual fluctuations and long-term trends of ice mass changes throughout the AIS. In this study, these two aspects were quantitatively analyzed through global positioning system (GPS) and gravity recovery and climate experiment/follow on (GRACE/GFO) over the past two decades. The nonlinear variation of GPS-inferred vertical land motion (VLM) and the influence of surface elastic load are of particular concern. The principal component analysis method is utilized to extract common mode signals from GPS time series, while correcting for various surface loads. The first principal components (PCs) accounted for 57.67%, 35.87%, 36.28%, and 36.03% of the total variances in the vertical components for GPS raw, atmospheric + nontidal oceanic (AO)-removed, AO + hydrographic model (AOH)-removed, and AO + GRACE/GFO-based load (AOG)-removed, respectively. Furthermore, the GPS vertical velocity, excluding the common mode component + AOG, yielded a median value of 0.13 mm/yr, which indicates that the retreat of ice mass has made a significant contribution to the GPS-observed VLM. In addition, the glacial isostatic adjustment (GIA) effect is found to play a key role in the large-scale VLM uplifting of the West AIS. After evaluating five different GIA models with GPS vertical velocity, we suggest that the ICE-6G_D model can more effectively correct GIA signals in GPS observations over Antarctica.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"4525-4535"},"PeriodicalIF":4.7,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10839024","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143105958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"C3DGS: Compressing 3D Gaussian Model for Surface Reconstruction of Large-Scale Scenes Based on Multiview UAV Images","authors":"Jiating Qian;Yiming Yan;Fengjiao Gao;Baoyu Ge;Maosheng Wei;Boyi Shangguan;Guangjun He","doi":"10.1109/JSTARS.2025.3529261","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3529261","url":null,"abstract":"Methods based on 3D Gaussian Splatting (3DGS) for surface reconstruction face challenges when applied to large-scale scenes captured by UAV. Because the number of 3D Gaussians increases dramatically, leading to significant computational requirement and limiting the fineness of surface reconstruction. To address this challenge, we propose C3DGS that compresses 3D Gaussian model and ensures the quality of surface reconstruction of large-scale scenes in the face of heavy computational costs. Our method quantifies the contribution of 3D Gaussians to the surface reconstruction and prunes redundant 3D Gaussians to reduce the computational requirement of the model. In addition, pruning 3D Gaussians inevitably incurs loss, and in order to guarantee as many details as possible in the surface reconstruction of a complex scene, we use a ray tracing volume rendering method that can better evaluate the opacity of 3D Gaussians. Furthermore, we introduce two regularization terms to enhance the geometric consistency of multiple views, thus improving the realism of surface reconstruction. Experiments show that our method outperforms other 3DGS-based surface reconstruction methods when facing large-scale scenes.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"4396-4409"},"PeriodicalIF":4.7,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10839501","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143105980","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}