International journal of applied earth observation and geoinformation : ITC journal最新文献_第5页

Landslide susceptibility assessment of the Wanzhou district: Merging landslide susceptibility modelling (LSM) with InSAR-derived ground deformation map 万州区滑坡易发性评估：滑坡易发性建模（LSM）与 InSAR 地面变形图的融合

IF 7.6 Q1 REMOTE SENSING

International journal of applied earth observation and geoinformation : ITC journal

Pub Date : 2025-02-01 DOI: 10.1016/j.jag.2025.104365

Chao Zhou , Lulu Gan , Ying Cao , Yue Wang , Samuele Segoni , Xuguo Shi , Mahdi Motagh , Ramesh P Singh

The prevalent catalog-based Landslide Susceptibility Modelling (LSM) operates under the assumption that future landslide occurrences mirror past and current patterns. Due to growing urban expansion and climate change, certain landslides follow new patterns of occurrence, disrupting the foundational assumption of catalog-based LSM and leading to constraints in the effectiveness of traditional susceptibility maps. Here, to address this problem, we proposed a method to produce more accurate and dynamic landslide susceptibility maps by coupling advanced Ensemble Machine Learning (EML) and Multi-Temporal Interferometric SAR (MT-InSAR). The Wanzhou District in Three Gorges Reservoir area of China is considered as the test site. The landslide catalog and multiple EML methods are used for the preparation of the preliminary susceptibility map. We have also compared and analyzed the impact of ensemble strategies (homogeneous and heterogeneous ensemble) and base-learners on the modelling performance. Subsequently, Sentinel-1 data from 2018 to 2020, analyzed using MT-InSAR approach, are used to map ground deformation rates. We outline the active slopes and deduce the relationship between the deformation of Matou landslide and triggering factors. The final susceptibility map is generated by coupling catalog-based susceptibility and ground deformation rate maps through an empirical assessment matrix. Our results show that the causal factors of distance to rivers, distance to faults, annual rainfall and distance to roads are basic parameters for landslide spatial development; Heterogeneous EML methods outperform the homogeneous, and the more base-learner types provide better performance. InSAR-acquired deformation rates corrected overestimation and underestimation errors in the landslide susceptibility map produced by catalog-based method. Our proposed method is capable of improving the accuracy and timeliness of susceptibility map, providing a useful instrument to better assess landslide risk scenarios in rapidly changing environments.

流行的基于目录的滑坡敏感性模型（LSM）是在假设未来的滑坡事件反映过去和现在的模式的情况下运行的。由于城市扩张和气候变化，某些滑坡遵循新的发生模式，破坏了基于目录的LSM的基本假设，并导致传统易感性图的有效性受到限制。在这里，为了解决这个问题，我们提出了一种方法，通过耦合先进的集成机器学习（EML）和多时相干涉SAR （MT-InSAR）来生成更准确和动态的滑坡易感性图。以三峡库区万州区为试验场。采用滑坡目录法和多重EML方法编制初步敏感性图。我们还比较和分析了集成策略（同质和异构集成）和基础学习器对建模性能的影响。随后，使用MT-InSAR方法分析2018年至2020年的Sentinel-1数据，用于绘制地面变形率。勾勒出活动边坡，推导出马头滑坡变形与诱发因素的关系。通过经验评估矩阵，将基于目录的敏感性图与地面变形率图耦合生成最终的敏感性图。结果表明：离河距离、离断层距离、年降雨量和离道路距离是影响滑坡空间发展的基本因素；异构EML方法优于同构EML方法，并且基础学习器类型越多，性能越好。insar获取的变形率修正了基于目录法生成的滑坡易感性图的高估和低估误差。该方法能够提高敏感性图的准确性和及时性，为快速变化环境下更好地评估滑坡风险情景提供了一种有用的工具。

{"title":"Landslide susceptibility assessment of the Wanzhou district: Merging landslide susceptibility modelling (LSM) with InSAR-derived ground deformation map","authors":"Chao Zhou , Lulu Gan , Ying Cao , Yue Wang , Samuele Segoni , Xuguo Shi , Mahdi Motagh , Ramesh P Singh","doi":"10.1016/j.jag.2025.104365","DOIUrl":"10.1016/j.jag.2025.104365","url":null,"abstract":"<div><div>The prevalent catalog-based Landslide Susceptibility Modelling (LSM) operates under the assumption that future landslide occurrences mirror past and current patterns. Due to growing urban expansion and climate change, certain landslides follow new patterns of occurrence, disrupting the foundational assumption of catalog-based LSM and leading to constraints in the effectiveness of traditional susceptibility maps. Here, to address this problem, we proposed a method to produce more accurate and dynamic landslide susceptibility maps by coupling advanced Ensemble Machine Learning (EML) and Multi-Temporal Interferometric SAR (MT-InSAR). The Wanzhou District in Three Gorges Reservoir area of China is considered as the test site. The landslide catalog and multiple EML methods are used for the preparation of the preliminary susceptibility map. We have also compared and analyzed the impact of ensemble strategies (homogeneous and heterogeneous ensemble) and base-learners on the modelling performance. Subsequently, Sentinel-1 data from 2018 to 2020, analyzed using MT-InSAR approach, are used to map ground deformation rates. We outline the active slopes and deduce the relationship between the deformation of Matou landslide and triggering factors. The final susceptibility map is generated by coupling catalog-based susceptibility and ground deformation rate maps through an empirical assessment matrix. Our results show that the causal factors of distance to rivers, distance to faults, annual rainfall and distance to roads are basic parameters for landslide spatial development; Heterogeneous EML methods outperform the homogeneous, and the more base-learner types provide better performance. InSAR-acquired deformation rates corrected overestimation and underestimation errors in the landslide susceptibility map produced by catalog-based method. Our proposed method is capable of improving the accuracy and timeliness of susceptibility map, providing a useful instrument to better assess landslide risk scenarios in rapidly changing environments.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"136 ","pages":"Article 104365"},"PeriodicalIF":7.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142990288","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Empirical methods to determine surface air temperature from satellite-retrieved data

IF 7.6 Q1 REMOTE SENSING

International journal of applied earth observation and geoinformation : ITC journal

Pub Date : 2025-02-01 DOI: 10.1016/j.jag.2025.104380

Joan Vedrí , Raquel Niclòs , Lluís Pérez-Planells , Enric Valor , Yolanda Luna , María José Estrela

Surface air temperature (SAT) is an essential climate variable (ECV). Models based on remote sensing data allow us to study SAT, without the need for a large network of meteorological stations. Therefore, it allows monitoring the climate in remote and extensive areas. Niclos et al. (2014) proposed parametric equations for the SAT retrieval over the Spanish Mediterranean basins. In this study, we evaluated those equations, but in a larger area and period of study. In addition, we proposed several linear regression models and nonlinear models based on decision tree methods, non-parametric methods and neuronal networks. These models relate SAT to land surface temperature, vegetation indexes and albedo from MODIS data. Moreover, meteorological reanalysis data, from ERA5-Land database, and geographical parameters were used. The accuracy of each model was evaluated against data from meteorological stations operated by AEMET in the Spanish Mediterranean basins, during the period 2021–2022. The equations of Niclos et al. (2014) obtained a robust root mean square error (RRMSE) of 3.1 K at daytime and 1.9 K at nighttime. For the linear regression models, the RRMSE decreased to 2.3 K (1.5 K) at daytime (nighttime). Finally, the nonlinear methods, in particular XGBoost model, showed an RRMSE of 1.5 K for daytime and 1.0 K at nighttime. Therefore, the comparison between methods showed that nonlinear models, in particular those based on decision tree methods, offered the best results in SAT retrieval in our study.

{"title":"Empirical methods to determine surface air temperature from satellite-retrieved data","authors":"Joan Vedrí , Raquel Niclòs , Lluís Pérez-Planells , Enric Valor , Yolanda Luna , María José Estrela","doi":"10.1016/j.jag.2025.104380","DOIUrl":"10.1016/j.jag.2025.104380","url":null,"abstract":"<div><div>Surface air temperature (SAT) is an essential climate variable (ECV). Models based on remote sensing data allow us to study SAT, without the need for a large network of meteorological stations. Therefore, it allows monitoring the climate in remote and extensive areas. Niclos et al. (2014) proposed parametric equations for the SAT retrieval over the Spanish Mediterranean basins. In this study, we evaluated those equations, but in a larger area and period of study. In addition, we proposed several linear regression models and nonlinear models based on decision tree methods, non-parametric methods and neuronal networks. These models relate SAT to land surface temperature, vegetation indexes and albedo from MODIS data. Moreover, meteorological reanalysis data, from ERA5-Land database, and geographical parameters were used. The accuracy of each model was evaluated against data from meteorological stations operated by AEMET in the Spanish Mediterranean basins, during the period 2021–2022. The equations of Niclos et al. (2014) obtained a robust root mean square error (RRMSE) of 3.1 K at daytime and 1.9 K at nighttime. For the linear regression models, the RRMSE decreased to 2.3 K (1.5 K) at daytime (nighttime). Finally, the nonlinear methods, in particular XGBoost model, showed an RRMSE of 1.5 K for daytime and 1.0 K at nighttime. Therefore, the comparison between methods showed that nonlinear models, in particular those based on decision tree methods, offered the best results in SAT retrieval in our study.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"136 ","pages":"Article 104380"},"PeriodicalIF":7.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143049967","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Tracking changes in wetlandscape properties of the Lake Winnipeg Watershed using Landsat inundation products (1984–2020)

IF 7.6 Q1 REMOTE SENSING

International journal of applied earth observation and geoinformation : ITC journal

Pub Date : 2025-02-01 DOI: 10.1016/j.jag.2025.104376

Forough Fendereski , Shizhou Ma , Sassan Mohammady , Christopher Spence , Charles G. Trick , Irena F. Creed

Wetlandscapes—hydrologically connected networks of wetlands—vary over time, causing changes in their provision of hydrological, biogeochemical, and ecological functions to landscapes. Here, we developed a method for mapping wetlands and extracting wetlandscape properties from Landsat-derived inundation data and applied this method to the Lake Winnipeg Watershed (LWW). We first mapped the annual (1984–2020) time series of inundated areas using a fusion of two Landsat-derived inundation products, Global Surface Water Extent (GSWE) and Dynamic Surface Water Extent (DSWE), finding that this fusion reduced omission errors from 17 % for GSWE and 18 % for DSWE to 8 % overall. We then used the inundated area maps to identify the topological structure of the wetlandscape, i.e., networks composed of nodes (representing wetlands) and their links (representing hydrological connectivity among wetlands). The time series of the wetlandscape properties (number, size, and connectivity of wetlands) showed coherence with a concurrent increase in precipitation over the watershed. The LWW is transitioning to a more extensive wetland area consisting of a greater number of larger wetlands with increased connections among them (p < 0.1). With Landsat-derived inundation products widely available globally, we suggest using the method developed here to analyze changes in wetlandscape properties in other regions worldwide.

{"title":"Tracking changes in wetlandscape properties of the Lake Winnipeg Watershed using Landsat inundation products (1984–2020)","authors":"Forough Fendereski , Shizhou Ma , Sassan Mohammady , Christopher Spence , Charles G. Trick , Irena F. Creed","doi":"10.1016/j.jag.2025.104376","DOIUrl":"10.1016/j.jag.2025.104376","url":null,"abstract":"<div><div>Wetlandscapes—hydrologically connected networks of wetlands—vary over time, causing changes in their provision of hydrological, biogeochemical, and ecological functions to landscapes. Here, we developed a method for mapping wetlands and extracting wetlandscape properties from Landsat-derived inundation data and applied this method to the Lake Winnipeg Watershed (LWW). We first mapped the annual (1984–2020) time series of inundated areas using a fusion of two Landsat-derived inundation products, Global Surface Water Extent (GSWE) and Dynamic Surface Water Extent (DSWE), finding that this fusion reduced omission errors from 17 % for GSWE and 18 % for DSWE to 8 % overall. We then used the inundated area maps to identify the topological structure of the wetlandscape, i.e., networks composed of nodes (representing wetlands) and their links (representing hydrological connectivity among wetlands). The time series of the wetlandscape properties (number, size, and connectivity of wetlands) showed coherence with a concurrent increase in precipitation over the watershed. The LWW is transitioning to a more extensive wetland area consisting of a greater number of larger wetlands with increased connections among them (<em>p</em> < 0.1). With Landsat-derived inundation products widely available globally, we suggest using the method developed here to analyze changes in wetlandscape properties in other regions worldwide.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"136 ","pages":"Article 104376"},"PeriodicalIF":7.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143050026","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Identification of standing dead trees in Robinia pseudoacacia plantations across China’s Loess Plateau using multiple deep learning models

IF 7.6 Q1 REMOTE SENSING

International journal of applied earth observation and geoinformation : ITC journal

Pub Date : 2025-02-01 DOI: 10.1016/j.jag.2025.104388

Li Zhang , Xiaodong Gao , Shuyi Zhou , Zhibo Zhang , Tianjie Zhao , Yaohui Cai , Xining Zhao

Drought-induced tree mortality has increasingly expanded worldwide under the influence of climate warming, with China’s Loess Plateau (CLP) emerging as a critical hotspot for such impacts. As one of the most active tree-planting regions globally, the CLP primarily aims to achieve soil and water conservation despite facing challenges such as limited rainfall and frequent extreme drought events. However, accurate identification of standing dead trees (SDTs) within plantations using remote sensing techniques remains underexplored, and the spatial distribution patterns of SDTs across the CLP are poorly understood. Therefore, this study leveraged unmanned aerial vehicle (UAV) remote sensing to capture high-resolution RGB images of Robinia pseudoacacia plantations. These images were then integrated with a comprehensive evaluation of multiple detection algorithms, including Faster R-CNN, EfficientDet, YOLOv4, YOLOv5, YOLOv8, YOLOv9, and a novel model, YOLOv9-ECA. Particularly, the YOLOv9-ECA was developed by incorporating the ECA module into key network layers to enhance channel dependency modeling and improve feature representation for SDTs detection. Its merit lies in adaptively reweighting feature channels, enabling efficient detection in resource-constrained environments. As expected, the YOLOv9-ECA model demonstrated significant advancements, achieving a detection speed of 123.5f/s, a mAP of 97.8%, and an F₁ score of 0.97, outperforming other models in both detection efficiency and accuracy. Subsequently, the model was employed to quantify the spatial distribution of SDTs across the CLP by estimating the number of dead trees per unit area. Results revealed an increasing trend in the number of dead trees per unit along decreasing precipitation gradients, emphasizing the vulnerability of Robinia pseudoacacia plantations in drier regions. Additionally, the number of dead trees per unit varied with slope aspect, with sunny slopes exhibiting the highest values and shady slopes the lowest. This study highlights the potential of YOLOv9-ECA as a powerful tool for the efficient detection of SDTs, offering insights for the sustainable management of Robinia pseudoacacia plantations on the CLP and holding potential applicability to similar environments globally.

{"title":"Identification of standing dead trees in Robinia pseudoacacia plantations across China’s Loess Plateau using multiple deep learning models","authors":"Li Zhang , Xiaodong Gao , Shuyi Zhou , Zhibo Zhang , Tianjie Zhao , Yaohui Cai , Xining Zhao","doi":"10.1016/j.jag.2025.104388","DOIUrl":"10.1016/j.jag.2025.104388","url":null,"abstract":"<div><div>Drought-induced tree mortality has increasingly expanded worldwide under the influence of climate warming, with China’s Loess Plateau (CLP) emerging as a critical hotspot for such impacts. As one of the most active tree-planting regions globally, the CLP primarily aims to achieve soil and water conservation despite facing challenges such as limited rainfall and frequent extreme drought events. However, accurate identification of standing dead trees (SDTs) within plantations using remote sensing techniques remains underexplored, and the spatial distribution patterns of SDTs across the CLP are poorly understood. Therefore, this study leveraged unmanned aerial vehicle (UAV) remote sensing to capture high-resolution RGB images of <em>Robinia pseudoacacia</em> plantations. These images were then integrated with a comprehensive evaluation of multiple detection algorithms, including Faster R-CNN, EfficientDet, YOLOv4, YOLOv5, YOLOv8, YOLOv9, and a novel model, YOLOv9-ECA. Particularly, the YOLOv9-ECA was developed by incorporating the ECA module into key network layers to enhance channel dependency modeling and improve feature representation for SDTs detection. Its merit lies in adaptively reweighting feature channels, enabling efficient detection in resource-constrained environments. As expected, the YOLOv9-ECA model demonstrated significant advancements, achieving a detection speed of 123.5f/s, a mAP of 97.8%, and an F<sub>1</sub> score of 0.97, outperforming other models in both detection efficiency and accuracy. Subsequently, the model was employed to quantify the spatial distribution of SDTs across the CLP by estimating the number of dead trees per unit area. Results revealed an increasing trend in the number of dead trees per unit along decreasing precipitation gradients, emphasizing the vulnerability of <em>Robinia pseudoacacia</em> plantations in drier regions. Additionally, the number of dead trees per unit varied with slope aspect, with sunny slopes exhibiting the highest values and shady slopes the lowest. This study highlights the potential of YOLOv9-ECA as a powerful tool for the efficient detection of SDTs, offering insights for the sustainable management of <em>Robinia pseudoacacia</em> plantations on the CLP and holding potential applicability to similar environments globally.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"136 ","pages":"Article 104388"},"PeriodicalIF":7.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143083302","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Identification of shrinkage patterns in Japan’s four major metropolitan areas based on nighttime light and population data

IF 7.6 Q1 REMOTE SENSING

International journal of applied earth observation and geoinformation : ITC journal

Pub Date : 2025-02-01 DOI: 10.1016/j.jag.2025.104391

Hao Zheng, Runsen Zhang

Urban shrinkage has become a critical global issue, influencing the sustainable development of cities across social, economic, and environmental dimensions. In Japan, which is characterized by an aging population and low birth rate, this phenomenon has now extended to metropolitan areas, presenting new challenges for urban sustainability. Although many studies have been conducted regarding urban decline in rural regions, the shrinkage dynamics within Japan’s major cities are poorly understood. Addressing this knowledge gap is crucial for devising targeted urban-planning strategies that ensure the long-term viability of urban areas. Here, we integrated Suomi National Polar-orbiting Partnership–Visible Infrared Imager Radiometer Suite nighttime light data with WorldPop population data to examine the patterns of urban shrinkage from 2012 to 2020 in Japan’s four largest metropolitan areas: Tokyo, Osaka, Nagoya, and Fukuoka. Using Theil–Sen median trend analysis and K-means clustering, we developed a method to quantify both shrinking and growing areas within these regions. It was found that Tokyo exhibited the highest urban vitality, with minimal shrinkage, whereas Nagoya and Osaka faced greater declines. Fukuoka displayed a distinct east–west spatial pattern of urban shrinkage. This study introduces the “triple V” theory, which evaluates urban vitality through the lenses of robustness and activity levels. Our analysis highlights the spatial complexities of urban shrinkage, emphasizing the importance of region-specific urban planning. By providing new insights obtained from a data-driven analysis, we offer a framework for policymakers to promote sustainable urban development in the face of demographic and spatial challenges.

{"title":"Identification of shrinkage patterns in Japan’s four major metropolitan areas based on nighttime light and population data","authors":"Hao Zheng, Runsen Zhang","doi":"10.1016/j.jag.2025.104391","DOIUrl":"10.1016/j.jag.2025.104391","url":null,"abstract":"<div><div>Urban shrinkage has become a critical global issue, influencing the sustainable development of cities across social, economic, and environmental dimensions. In Japan, which is characterized by an aging population and low birth rate, this phenomenon has now extended to metropolitan areas, presenting new challenges for urban sustainability. Although many studies have been conducted regarding urban decline in rural regions, the shrinkage dynamics within Japan’s major cities are poorly understood. Addressing this knowledge gap is crucial for devising targeted urban-planning strategies that ensure the long-term viability of urban areas. Here, we integrated Suomi National Polar-orbiting Partnership–Visible Infrared Imager Radiometer Suite nighttime light data with WorldPop population data to examine the patterns of urban shrinkage from 2012 to 2020 in Japan’s four largest metropolitan areas: Tokyo, Osaka, Nagoya, and Fukuoka. Using Theil–Sen median trend analysis and K-means clustering, we developed a method to quantify both shrinking and growing areas within these regions. It was found that Tokyo exhibited the highest urban vitality, with minimal shrinkage, whereas Nagoya and Osaka faced greater declines. Fukuoka displayed a distinct east–west spatial pattern of urban shrinkage. This study introduces the “triple V” theory, which evaluates urban vitality through the lenses of robustness and activity levels. Our analysis highlights the spatial complexities of urban shrinkage, emphasizing the importance of region-specific urban planning. By providing new insights obtained from a data-driven analysis, we offer a framework for policymakers to promote sustainable urban development in the face of demographic and spatial challenges.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"136 ","pages":"Article 104391"},"PeriodicalIF":7.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143137371","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

OptiViewNeRF: Optimizing 3D reconstruction via batch view selection and scene uncertainty in Neural Radiance Fields OptiViewNeRF：在神经辐射场中通过批量视图选择和场景不确定性优化3D重建

IF 7.6 Q1 REMOTE SENSING

International journal of applied earth observation and geoinformation : ITC journal

Pub Date : 2025-02-01 DOI: 10.1016/j.jag.2024.104306

You Li , Rui Li , Ziwei Li , Renzhong Guo , Shengjun Tang

In situations with a limited number of posed images, choosing the most suitable viewpoints becomes crucial for accurate Neural Radiance Fields (NeRF) modeling. Current approaches for view selection often rely on heuristic methods or are computationally intensive. To address these challenges, we introduce a new framework, OptiViewNeRF, which leverages scene uncertainty to guide the view selection process. Initially, an uncertainty estimation model of the entire scene is developed based on a preliminary NeRF model. This model then informs the selection of new perception viewpoints using a batch view selection strategy, allowing the entire process to be completed in a single iteration. By selecting viewpoints that provide informative data, this approach improves novel view synthesis results and accurately reconstructs 3D scenes. Experimental results on two selected datasets show that the proposed method effectively identifies informative viewpoints, resulting in more accurate scene reconstructions compared to baseline and state-of-the-art methods.

在摆姿势图像数量有限的情况下，选择最合适的视点对于准确的神经辐射场（NeRF）建模至关重要。当前的视图选择方法通常依赖于启发式方法或计算密集型方法。为了应对这些挑战，我们引入了一个新的框架，OptiViewNeRF，它利用场景的不确定性来指导视图选择过程。首先，在初步NeRF模型的基础上，建立了整个场景的不确定性估计模型。然后，该模型使用批处理视图选择策略通知新感知视点的选择，从而允许在单个迭代中完成整个过程。通过选择提供信息数据的视点，该方法改进了新的视图合成结果，并准确地重建了3D场景。在两个选定的数据集上的实验结果表明，该方法有效地识别了信息视角，与基线和当前的方法相比，产生了更准确的场景重建。

{"title":"OptiViewNeRF: Optimizing 3D reconstruction via batch view selection and scene uncertainty in Neural Radiance Fields","authors":"You Li , Rui Li , Ziwei Li , Renzhong Guo , Shengjun Tang","doi":"10.1016/j.jag.2024.104306","DOIUrl":"10.1016/j.jag.2024.104306","url":null,"abstract":"<div><div>In situations with a limited number of posed images, choosing the most suitable viewpoints becomes crucial for accurate Neural Radiance Fields (NeRF) modeling. Current approaches for view selection often rely on heuristic methods or are computationally intensive. To address these challenges, we introduce a new framework, OptiViewNeRF, which leverages scene uncertainty to guide the view selection process. Initially, an uncertainty estimation model of the entire scene is developed based on a preliminary NeRF model. This model then informs the selection of new perception viewpoints using a batch view selection strategy, allowing the entire process to be completed in a single iteration. By selecting viewpoints that provide informative data, this approach improves novel view synthesis results and accurately reconstructs 3D scenes. Experimental results on two selected datasets show that the proposed method effectively identifies informative viewpoints, resulting in more accurate scene reconstructions compared to baseline and state-of-the-art methods.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"136 ","pages":"Article 104306"},"PeriodicalIF":7.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142975204","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

3D-UMamba: 3D U-Net with state space model for semantic segmentation of multi-source LiDAR point clouds

IF 7.6 Q1 REMOTE SENSING

International journal of applied earth observation and geoinformation : ITC journal

Pub Date : 2025-02-01 DOI: 10.1016/j.jag.2025.104401

Dening Lu , Linlin Xu , Jun Zhou , Kyle Gao , Zheng Gong , Dedong Zhang

Segmentation of point clouds is foundational to numerous remote sensing applications. Recently, the development of Transformers has further improved segmentation techniques thanks to their great long-range context modeling capability. However, Transformers have quadratic complexity in inference time and memory, which both limits the input size and poses a strict hardware requirement. This paper presents a novel 3D-UMamba network with linear complexity, which is the earliest to introduce the Selective State Space Model (i.e., Mamba) to multi-source LiDAR point cloud processing. 3D-UMamba integrates Mamba into the classic U-Net architecture, presenting outstanding global context modeling with high efficiency and achieving an effective combination of local and global information. In addition, we propose a simple yet efficient 3D-token serialization approach (Voxel-based Token Serialization, i.e., VTS) for Mamba, where the Bi-Scanning strategy enables the model to collect features from all input points in different directions effectively. The performance of 3D-UMamba on three challenging LiDAR point cloud datasets (airborne MultiSpectral LiDAR (MS-LiDAR), aerial DALES, and vehicle-mounted Toronto-3D) demonstrated its superiority in multi-source LiDAR point cloud semantic segmentation, as well as the strong adaptability of Mamba to different types of LiDAR data, exceeding current state-of-the-art models. Ablation studies demonstrated the higher efficiency and lower memory costs of 3D-UMamba than its Transformer-based counterparts.

{"title":"3D-UMamba: 3D U-Net with state space model for semantic segmentation of multi-source LiDAR point clouds","authors":"Dening Lu , Linlin Xu , Jun Zhou , Kyle Gao , Zheng Gong , Dedong Zhang","doi":"10.1016/j.jag.2025.104401","DOIUrl":"10.1016/j.jag.2025.104401","url":null,"abstract":"<div><div>Segmentation of point clouds is foundational to numerous remote sensing applications. Recently, the development of Transformers has further improved segmentation techniques thanks to their great long-range context modeling capability. However, Transformers have quadratic complexity in inference time and memory, which both limits the input size and poses a strict hardware requirement. This paper presents a novel 3D-UMamba network with linear complexity, which is the earliest to introduce the Selective State Space Model (i.e., Mamba) to multi-source LiDAR point cloud processing. 3D-UMamba integrates Mamba into the classic U-Net architecture, presenting outstanding global context modeling with high efficiency and achieving an effective combination of local and global information. In addition, we propose a simple yet efficient 3D-token serialization approach (Voxel-based Token Serialization, i.e., VTS) for Mamba, where the Bi-Scanning strategy enables the model to collect features from all input points in different directions effectively. The performance of 3D-UMamba on three challenging LiDAR point cloud datasets (airborne MultiSpectral LiDAR (MS-LiDAR), aerial DALES, and vehicle-mounted Toronto-3D) demonstrated its superiority in multi-source LiDAR point cloud semantic segmentation, as well as the strong adaptability of Mamba to different types of LiDAR data, exceeding current state-of-the-art models. Ablation studies demonstrated the higher efficiency and lower memory costs of 3D-UMamba than its Transformer-based counterparts.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"136 ","pages":"Article 104401"},"PeriodicalIF":7.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143394677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A spatiotemporal framework to assess the bio-geomorphic interplay of saltmarsh vegetation and tidal emergence (Western Scheldt estuary) 盐沼植被与潮汐相互作用的时空框架研究（西谢尔德河河口）

IF 7.6 Q1 REMOTE SENSING

International journal of applied earth observation and geoinformation : ITC journal

Pub Date : 2025-02-01 DOI: 10.1016/j.jag.2024.104337

Jing Feng , Tim J. Grandjean , Johan van de Koppel , Daphne van der Wal

Sea level changes will significantly drive hydrodynamic, morphological, and ecological development of estuaries. However, the interplay of geomorphology and vegetation at estuary scales remains unclear. To better understand this process, we take the Western Scheldt estuary in the Netherlands as an example to reveal the link between changes in emersion duration and vegetation dynamics in the period 1993–2016. We found that tidal flats in the Western Scheldt become steeper—higher intertidal areas increased in elevation and emersion duration, whereas the low-lying edges of tidal flats experienced a decrease in elevation and emersion duration. We found that longer emersion duration was associated with increased plant diversity and cover. Furthermore, we detected the unique spatiotemporal response patterns of four abundant plant species to geomorphological variations. Our study suggests that on a large estuary scale, geomorphological changes are coupled to the richness and cover of plant communities, and that potential changes in relative sea level can induce structural modifications of the plant communities. It also emphasizes the importance of assessing the potential effects of localized relative sea level changes while considering all aspects of natural processes and direct and indirect human influences. Our study provides a framework to assess the bio-geomorphic processes in a spatially explicit way.

海平面的变化将对河口的水动力、形态和生态发展产生重要影响。然而，在河口尺度上，地貌与植被的相互作用尚不清楚。为了更好地理解这一过程，我们以荷兰西舍尔德河河口为例，揭示了1993-2016年期间海蚀期变化与植被动态之间的联系。研究发现，西斯海尔德河潮滩坡度陡，潮间带高程增加，潮间带高程增加，潮间带低洼边缘高程减少，潮间带高程减少。我们发现，较长的重现时间与增加的植物多样性和覆盖有关。此外，我们还检测了四种丰富的植物物种对地貌变化的独特时空响应模式。研究表明，在大河口尺度上，地貌变化与植物群落的丰富度和覆盖度是耦合的，相对海平面的潜在变化可以引起植物群落的结构改变。它还强调评估局部相对海平面变化的潜在影响的重要性，同时考虑到自然过程的所有方面以及直接和间接的人类影响。我们的研究提供了一个框架，以空间明确的方式评估生物地貌过程。

{"title":"A spatiotemporal framework to assess the bio-geomorphic interplay of saltmarsh vegetation and tidal emergence (Western Scheldt estuary)","authors":"Jing Feng , Tim J. Grandjean , Johan van de Koppel , Daphne van der Wal","doi":"10.1016/j.jag.2024.104337","DOIUrl":"10.1016/j.jag.2024.104337","url":null,"abstract":"<div><div>Sea level changes will significantly drive hydrodynamic, morphological, and ecological development of estuaries. However, the interplay of geomorphology and vegetation at estuary scales remains unclear. To better understand this process, we take the Western Scheldt estuary in the Netherlands as an example to reveal the link between changes in emersion duration and vegetation dynamics in the period 1993–2016. We found that tidal flats in the Western Scheldt become steeper—higher intertidal areas increased in elevation and emersion duration, whereas the low-lying edges of tidal flats experienced a decrease in elevation and emersion duration. We found that longer emersion duration was associated with increased plant diversity and cover. Furthermore, we detected the unique spatiotemporal response patterns of four abundant plant species to geomorphological variations. Our study suggests that on a large estuary scale, geomorphological changes are coupled to the richness and cover of plant communities, and that potential changes in relative sea level can induce structural modifications of the plant communities. It also emphasizes the importance of assessing the potential effects of localized relative sea level changes while considering all aspects of natural processes and direct and indirect human influences. Our study provides a framework to assess the bio-geomorphic processes in a spatially explicit way.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"136 ","pages":"Article 104337"},"PeriodicalIF":7.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142874795","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Deep learning change detection techniques for optical remote sensing imagery: Status, perspectives and challenges 光学遥感图像的深度学习变化检测技术：现状、展望和挑战

IF 7.6 Q1 REMOTE SENSING

International journal of applied earth observation and geoinformation : ITC journal

Pub Date : 2025-02-01 DOI: 10.1016/j.jag.2024.104282

Daifeng Peng , Xuelian Liu , Yongjun Zhang , Haiyan Guan , Yansheng Li , Lorenzo Bruzzone

Change detection (CD) aims to compare and analyze images of identical geographic areas but different dates, whereby revealing spatio-temporal change patterns of Earth’s surface. With the implementation of the High-Resolution Earth Observation Project, an integrated sky-to-ground observation system has been continuously developed and improved. The accumulation of massive multi-modal, multi-angle, and multi-resolution remote sensing data have greatly enriched the CD data sources. Among them, high-resolution optical remote sensing images contain abundant spatial detail information, making it possible to interpret fine-grained scenes and greatly expand the application breadth and depth of CD. Generally, traditional optical remote sensing CD methods are cumbersome in steps and have a low level of automation. In contrast, artificial intelligence (AI) based CD methods possess powerful feature extraction and non-linear modeling capabilities, thereby gaining advantages that traditional methods cannot match. As a result, they have become the mainstream approaches in the field of CD. This review article systematically summarizes the datasets, theories, and methods of CD for optical remote sensing image. It provides a comprehensive analysis of AI-based CD algorithms based on deep learning paradigms from the perspectives of algorithm granularity. In-depth analysis of the performance of typical algorithms are further conducted. Finally, we summarize the challenges and trends of the CD algorithms in the AI era, aiming to provide important guidelines and insights for relevant researchers.

变化检测（Change detection， CD）旨在对相同地理区域不同日期的图像进行比较分析，从而揭示地球表面的时空变化规律。随着高分辨率对地观测工程的实施，对地综合观测系统不断发展完善。大量多模态、多角度、多分辨率遥感数据的积累，极大地丰富了遥感数据的来源。其中，高分辨率光学遥感影像包含了丰富的空间细节信息，使得对细粒度场景的解读成为可能，极大地拓展了CD的应用广度和深度。传统的光学遥感CD方法一般步骤繁琐，自动化程度较低。而基于人工智能（AI）的CD方法具有强大的特征提取和非线性建模能力，具有传统方法无法比拟的优势。本文系统地综述了光学遥感图像的数据集、理论和方法。从算法粒度的角度全面分析了基于深度学习范式的基于ai的CD算法。进一步深入分析了典型算法的性能。最后，我们总结了人工智能时代CD算法面临的挑战和趋势，旨在为相关研究人员提供重要的指导和见解。

{"title":"Deep learning change detection techniques for optical remote sensing imagery: Status, perspectives and challenges","authors":"Daifeng Peng , Xuelian Liu , Yongjun Zhang , Haiyan Guan , Yansheng Li , Lorenzo Bruzzone","doi":"10.1016/j.jag.2024.104282","DOIUrl":"10.1016/j.jag.2024.104282","url":null,"abstract":"<div><div>Change detection (CD) aims to compare and analyze images of identical geographic areas but different dates, whereby revealing spatio-temporal change patterns of Earth’s surface. With the implementation of the High-Resolution Earth Observation Project, an integrated sky-to-ground observation system has been continuously developed and improved. The accumulation of massive multi-modal, multi-angle, and multi-resolution remote sensing data have greatly enriched the CD data sources. Among them, high-resolution optical remote sensing images contain abundant spatial detail information, making it possible to interpret fine-grained scenes and greatly expand the application breadth and depth of CD. Generally, traditional optical remote sensing CD methods are cumbersome in steps and have a low level of automation. In contrast, artificial intelligence (AI) based CD methods possess powerful feature extraction and non-linear modeling capabilities, thereby gaining advantages that traditional methods cannot match. As a result, they have become the mainstream approaches in the field of CD. This review article systematically summarizes the datasets, theories, and methods of CD for optical remote sensing image. It provides a comprehensive analysis of AI-based CD algorithms based on deep learning paradigms from the perspectives of algorithm granularity. In-depth analysis of the performance of typical algorithms are further conducted. Finally, we summarize the challenges and trends of the CD algorithms in the AI era, aiming to provide important guidelines and insights for relevant researchers.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"136 ","pages":"Article 104282"},"PeriodicalIF":7.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142874860","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Spatiotemporal-grained quantitative assessment of construction-induced deformation along the MTR in Hong Kong using MT-InSAR and iterative STL-based subsidence ratio analysis 利用MT-InSAR和基于迭代stl的沉降比分析对香港地铁沿线施工引起的变形进行时空粒度定量评估

IF 7.6 Q1 REMOTE SENSING

International journal of applied earth observation and geoinformation : ITC journal

Pub Date : 2025-02-01 DOI: 10.1016/j.jag.2024.104342

Jiayuan Zhang , Yuhao Liu , Bochen Zhang , Siting Xiong , Chisheng Wang , Songbo Wu , Wu Zhu

Multi-temporal synthetic aperture radar interferometry (MT-InSAR) offers unique advantages in monitoring ground deformation and structural stability along the metro lines. However, a vast number of complex deformation points, millions and even more, can be derived from InSAR making it challenging to identify the deformation hotspot in time series automatically. This paper proposes a novel method for quantitatively assessing the MT-InSAR-derived deformation results. We first introduce an iterative seasonal trend decomposition using loess (STL) method to confirm the optimal period for separating seasonal components from the displacement time series. Then, an absolute differences detector with rolling windows is proposed to quantify the subsidence ratio within the time series and allow deformation hotspots to be more visible. To validate the effectiveness of the proposed method, 468 scenes of Sentinel-1A ascending images from Jun. 2015 to Nov. 2023 over the Hong Kong Mass Transit Railway (MTR) are adopted. The results indicate that 99.2% of areas are relatively stable with the displacement velocity ranging from −2 mm/year to 2 mm/year, and 84% of the study area remained a subsidence ratio below 0.3, except for localized hotspots that exhibited either short or long-term subsidence trends. The findings of this study indicate that multiple deformation hotspots were identified at the intersections of several metro lines in the Kowloon Peninsula and along the Island line. In addition to the displacement velocity from the conventional MT-InSAR, the overall and annual subsidence ratios have been demonstrated to be useful indicators for quantitative assessment of the construction-induced deformation.

多时相合成孔径雷达干涉测量技术（MT-InSAR）在地铁沿线的地面变形和结构稳定性监测方面具有独特的优势。然而，InSAR可以获得大量复杂的变形点，数百万甚至更多，这给自动识别时间序列中的变形热点带来了挑战。本文提出了一种定量评估mt - insar衍生变形结果的新方法。首先采用黄土（STL）方法进行迭代季节趋势分解，确定从位移时间序列中分离季节分量的最优周期。然后，提出了一种带滚动窗的绝对差值检测器，以量化时间序列内的沉降比，并使变形热点更加明显。为了验证该方法的有效性，采用了2015年6月至2023年11月在香港地下铁路（MTR）上拍摄的468幅Sentinel-1A上升图像。结果表明：99.2%的区域相对稳定，位移速度在−2 mm/年~ 2 mm/年之间；除局部热点地区表现出短期或长期沉降趋势外，84%的区域沉降率保持在0.3以下；研究结果表明，在九龙半岛和港岛线沿线多条地铁线路的交汇处发现了多个变形热点。除了常规MT-InSAR的位移速度外，总体沉降比和年沉降比已被证明是定量评估施工引起变形的有用指标。

{"title":"Spatiotemporal-grained quantitative assessment of construction-induced deformation along the MTR in Hong Kong using MT-InSAR and iterative STL-based subsidence ratio analysis","authors":"Jiayuan Zhang , Yuhao Liu , Bochen Zhang , Siting Xiong , Chisheng Wang , Songbo Wu , Wu Zhu","doi":"10.1016/j.jag.2024.104342","DOIUrl":"10.1016/j.jag.2024.104342","url":null,"abstract":"<div><div>Multi-temporal synthetic aperture radar interferometry (MT-InSAR) offers unique advantages in monitoring ground deformation and structural stability along the metro lines. However, a vast number of complex deformation points, millions and even more, can be derived from InSAR making it challenging to identify the deformation hotspot in time series automatically. This paper proposes a novel method for quantitatively assessing the MT-InSAR-derived deformation results. We first introduce an iterative seasonal trend decomposition using loess (STL) method to confirm the optimal period for separating seasonal components from the displacement time series. Then, an absolute differences detector with rolling windows is proposed to quantify the subsidence ratio within the time series and allow deformation hotspots to be more visible. To validate the effectiveness of the proposed method, 468 scenes of Sentinel-1A ascending images from Jun. 2015 to Nov. 2023 over the Hong Kong Mass Transit Railway (MTR) are adopted. The results indicate that 99.2% of areas are relatively stable with the displacement velocity ranging from −2 mm/year to 2 mm/year, and 84% of the study area remained a subsidence ratio below 0.3, except for localized hotspots that exhibited either short or long-term subsidence trends. The findings of this study indicate that multiple deformation hotspots were identified at the intersections of several metro lines in the Kowloon Peninsula and along the Island line. In addition to the displacement velocity from the conventional MT-InSAR, the overall and annual subsidence ratios have been demonstrated to be useful indicators for quantitative assessment of the construction-induced deformation.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"136 ","pages":"Article 104342"},"PeriodicalIF":7.6,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142929502","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0