ISPRS Journal of Photogrammetry and Remote Sensing最新文献_第5页

Monitoring snow cover dynamics at 30-m resolution in higher latitude regions using Harmonized Landsat Sentinel-2 利用Harmonized Landsat Sentinel-2监测高纬度地区30米分辨率积雪动态

IF 12.2 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2026-03-01 Epub Date: 2026-01-31 DOI: 10.1016/j.isprsjprs.2026.01.032

Mitchell T. Bonney, Yu Zhang

Snow is an essential climate variable that is important for hydrology, climate, soil temperature and permafrost, vegetation, animal habitat, and socioeconomics. Wide-area snow cover dynamics (SCD), including the start and end of snow cover, are generally monitored by satellites with coarse spatial resolutions (250–1000 m) and high temporal (daily) resolutions. Higher spatial resolution (HSR) monitoring (10–30 m) has been limited to small areas because of computational constraints and infrequent cloud-free observations. Here, we develop a new method to map wide-area HSR SCD (snow start date, end date, length, periods, status) by leveraging the recently released Harmonized Landsat Sentinel-2 (HLS) v2.0, which has a 2–3-day revisit at 30-m resolution. The method is built around SpatialTemporal Asset Catalogs (STACs) and open-source Python tools. We utilize tiled datacubes, snow classification, and a model involving implausibility checking, cleaning, and finding peaks in data with gaps due to orbit frequencies and clouds. We demonstrate SCD mapping and validation across Canada’s Hudson Bay Lowlands (HBL) and an area in northern Alaska for each snow-year from 2018 to 2019 to 2023–2024 and multi-year composites (2018–2024). We also provide timing uncertainties and a quality metric for all pixels. Performance is best for snow end date, having strong relationships with both visually interpreted SCD from primarily very high-resolution imagery and measured local-scale snow depth. The combination of lower cloud cover and lower solar zenith angles during melt periods leads to lower uncertainties for snow end date compared to start date and length. Performance is better for all metrics at higher latitudes (e.g., northern Alaska), where satellite observations are more frequent due to increased orbit overlap. Although we have only completed validation for the HBL, Canada-wide products using this methodology are available publicly as STACs on the CCMEO Data Cube and will continue to be updated. Addition validation across Canada and methodology improvements are ongoing.

雪是一个重要的气候变量，对水文、气候、土壤温度和永久冻土、植被、动物栖息地和社会经济都很重要。广域积雪动态（SCD），包括积雪的开始和结束，通常由卫星监测，具有粗空间分辨率（250-1000 m）和高时间分辨率（日）。由于计算限制和不频繁的无云观测，高空间分辨率（HSR）监测（10-30米）仅限于小区域。在这里，我们利用最近发布的Harmonized Landsat Sentinel-2 (HLS) v2.0开发了一种新的方法来绘制广域高铁SCD（雪开始日期，结束日期，长度，周期，状态），该v2.0具有2 - 3天的30米分辨率重访。该方法是围绕时空资产目录（STACs）和开源Python工具构建的。我们使用了平铺数据库、雪分类和一个模型，该模型涉及不可信检查、清理和查找由于轨道频率和云而存在间隙的数据中的峰值。我们展示了加拿大哈德逊湾低地（HBL）和阿拉斯加北部地区的SCD制图和验证，包括2018年至2019年、2023年至2024年的每个雪年和多年合成（2018年至2024年）。我们还为所有像素提供了时间不确定性和质量度量。性能最好的是降雪结束日期，与主要是非常高分辨率图像的视觉解释SCD和测量的局地尺度雪深有很强的关系。融雪期较低的云量和较低的太阳天顶角相结合，导致雪结束日期的不确定性低于开始日期和长度。在高纬度地区（例如阿拉斯加北部），由于轨道重叠增加，卫星观测更加频繁，因此所有指标的性能都更好。虽然我们只完成了HBL的验证，但使用这种方法的加拿大范围内的产品在CCMEO数据立方上作为stac公开可用，并将继续更新。加拿大各地的附加验证和方法改进正在进行中。

{"title":"Monitoring snow cover dynamics at 30-m resolution in higher latitude regions using Harmonized Landsat Sentinel-2","authors":"Mitchell T. Bonney, Yu Zhang","doi":"10.1016/j.isprsjprs.2026.01.032","DOIUrl":"10.1016/j.isprsjprs.2026.01.032","url":null,"abstract":"<div><div>Snow is an essential climate variable that is important for hydrology, climate, soil temperature and permafrost, vegetation, animal habitat, and socioeconomics. Wide-area snow cover dynamics (SCD), including the start and end of snow cover, are generally monitored by satellites with coarse spatial resolutions (250–1000 m) and high temporal (daily) resolutions. Higher spatial resolution (HSR) monitoring (10–30 m) has been limited to small areas because of computational constraints and infrequent cloud-free observations. Here, we develop a new method to map wide-area HSR SCD (snow start date, end date, length, periods, status) by leveraging the recently released Harmonized Landsat Sentinel-2 (HLS) v2.0, which has a 2–3-day revisit at 30-m resolution. The method is built around SpatialTemporal Asset Catalogs (STACs) and open-source Python tools. We utilize tiled datacubes, snow classification, and a model involving implausibility checking, cleaning, and finding peaks in data with gaps due to orbit frequencies and clouds. We demonstrate SCD mapping and validation across Canada’s Hudson Bay Lowlands (HBL) and an area in northern Alaska for each snow-year from 2018 to 2019 to 2023–2024 and multi-year composites (2018–2024). We also provide timing uncertainties and a quality metric for all pixels. Performance is best for snow end date, having strong relationships with both visually interpreted SCD from primarily very high-resolution imagery and measured local-scale snow depth. The combination of lower cloud cover and lower solar zenith angles during melt periods leads to lower uncertainties for snow end date compared to start date and length. Performance is better for all metrics at higher latitudes (e.g., northern Alaska), where satellite observations are more frequent due to increased orbit overlap. Although we have only completed validation for the HBL, Canada-wide products using this methodology are available publicly as STACs on the CCMEO Data Cube and will continue to be updated. Addition validation across Canada and methodology improvements are ongoing.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"233 ","pages":"Pages 360-382"},"PeriodicalIF":12.2,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079876","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}

引用次数: 0

Identifying green leaf and leaf phenology of large trees and forests by time series PlanetScope and Sentinel-2 images and the chlorophyll and green leaf indicator (CGLI) 利用PlanetScope和Sentinel-2时间序列影像及叶绿素和绿叶指示剂（CGLI）识别大型树木和森林的绿叶和叶物候

IF 12.2 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2026-03-01 Epub Date: 2026-01-22 DOI: 10.1016/j.isprsjprs.2026.01.027

Baihong Pan , Xiangming Xiao , Li Pan , Andrew D Richardson , Yujie Liu , Yuan Yao , Cheng Meng , Yanhua Xie , Chenchen Zhang , Yuanwei Qin

Plant phenology serves as a vital indicator of plant’s response to climate variation and change. To date, our knowledge and data products of plant leaf phenology at the scales of large trees and forest stand are very limited, in part due to the lack of time series image data at very high spatial resolution (VHSR, meters). Here, we investigated surface reflectance (BLUE, GREEN, RED) and vegetation indices over a large cottonwood tree, using images from PlanetScope (daily, 3-m) and Sentinel-2A/B (5-day, 10-m) in 2023 and in-situ field photos. At the leaf scale, a green leaf has a spectral signature of BLUE < GREEN > RED, as chlorophyll pigment absorbs more blue and red light than green light, which is named as chlorophyll and green leaf indicator (CGLI); and a dead leaf has BLUE < GREEN < RED. At the tree scale, tree with only branches and trunk (no green leaves) has BLUE < GREEN < RED, while tree with green leaves has BLUE < GREEN > RED. We evaluated the start of season (SOS) and end of season (EOS) of the cottonwood tree, derived from (1) vegetation index (VI) data with three methods (VI-slope-, VI-ratio-, and VI-threshold-based methods) and (2) surface reflectance data with CGLI-based method. To evaluate broader applicability of the CGLI-based method, we applied the same workflow to five deciduous broadleaf forest sites within the National Ecological Observatory Network, equipped with PhenoCam. At these five sites, we compared phenology metrics (SOS, EOS) derived from VI- and CGLI-based methods with reference dates derived from PhenoCam Green Chromatic Coordinate (GCC) data. Results show that the CGLI-based method, which classifies each observation as either green leaf or non-green leaf/canopy (binary), is simple and effective in delineating leaf/canopy dynamics and phenology metrics. These findings provide a foundation for monitoring leaf phenology of large trees using satellite data.

植物物候是反映植物对气候变化响应的重要指标。迄今为止，由于缺乏非常高空间分辨率（VHSR，米）的时间序列图像数据，我们在大树和林分尺度上的植物叶物候知识和数据产品非常有限。在这里，我们利用PlanetScope（每日，3米）和Sentinel-2A/B（5天，10米）在2023年的图像和现场照片，研究了一棵大型棉杨树的表面反射率（蓝色，绿色，红色）和植被指数。在叶片尺度上，绿叶的光谱特征为BLUE <； green >； RED，这是因为叶绿素色素吸收的蓝光和红光比绿光多，称为叶绿素和绿叶指示剂（CGLI）；而枯叶则是蓝<；绿<；红。在树的尺度上，只有树枝和树干（没有绿叶）的树是BLUE <； green <； RED，有绿叶的树是BLUE <； green >； RED。本文对杨树的季初（SOS）和季末（EOS）数据进行了评价，该数据来源于：(1)基于植被指数（VI）的三种方法（VI-slope- based、VI-ratio- based和VI-threshold-based方法）和(2)基于cgi的地表反射率数据。为了评估基于cgi方法的更广泛适用性，我们将相同的工作流程应用于国家生态观测站网络内的五个落叶阔叶林站点，并配备了PhenoCam。在这五个地点，我们将基于VI和cgi方法得出的物候指标（SOS， EOS）与来自PhenoCam Green Chromatic Coordinate （GCC）数据的参考日期进行了比较。结果表明，基于cgi的方法将每个观测值分为绿叶或非绿叶/冠层（二元），在描述叶/冠层动态和物候指标方面简单有效。这些发现为利用卫星数据监测大型树木叶片物候提供了基础。

{"title":"Identifying green leaf and leaf phenology of large trees and forests by time series PlanetScope and Sentinel-2 images and the chlorophyll and green leaf indicator (CGLI)","authors":"Baihong Pan , Xiangming Xiao , Li Pan , Andrew D Richardson , Yujie Liu , Yuan Yao , Cheng Meng , Yanhua Xie , Chenchen Zhang , Yuanwei Qin","doi":"10.1016/j.isprsjprs.2026.01.027","DOIUrl":"10.1016/j.isprsjprs.2026.01.027","url":null,"abstract":"<div><div>Plant phenology serves as a vital indicator of plant’s response to climate variation and change. To date, our knowledge and data products of plant leaf phenology at the scales of large trees and forest stand are very limited, in part due to the lack of time series image data at very high spatial resolution (VHSR, meters). Here, we investigated surface reflectance (BLUE, GREEN, RED) and vegetation indices over a large cottonwood tree, using images from PlanetScope (daily, 3-m) and Sentinel-2A/B (5-day, 10-m) in 2023 and in-situ field photos. At the leaf scale, a green leaf has a spectral signature of BLUE < GREEN > RED, as chlorophyll pigment absorbs more blue and red light than green light, which is named as chlorophyll and green leaf indicator (CGLI); and a dead leaf has BLUE < GREEN < RED. At the tree scale, tree with only branches and trunk (no green leaves) has BLUE < GREEN < RED, while tree with green leaves has BLUE < GREEN > RED. We evaluated the start of season (SOS) and end of season (EOS) of the cottonwood tree, derived from (1) vegetation index (VI) data with three methods (VI-slope-, VI-ratio-, and VI-threshold-based methods) and (2) surface reflectance data with CGLI-based method. To evaluate broader applicability of the CGLI-based method, we applied the same workflow to five deciduous broadleaf forest sites within the National Ecological Observatory Network, equipped with PhenoCam. At these five sites, we compared phenology metrics (SOS, EOS) derived from VI- and CGLI-based methods with reference dates derived from PhenoCam Green Chromatic Coordinate (GCC) data. Results show that the CGLI-based method, which classifies each observation as either green leaf or non-green leaf/canopy (binary), is simple and effective in delineating leaf/canopy dynamics and phenology metrics. These findings provide a foundation for monitoring leaf phenology of large trees using satellite data.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"233 ","pages":"Pages 104-125"},"PeriodicalIF":12.2,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146033660","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}

引用次数: 0

AnchorReF: A novel anchor-based visual re-localization framework aided by multi-sensor data fusion 基于多传感器数据融合的基于锚点的视觉再定位框架

IF 12.2 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2026-03-01 Epub Date: 2026-01-16 DOI: 10.1016/j.isprsjprs.2026.01.019

Hao Wu , Yu Ran , Xiaoxiang Zhang, Xinying Luo, Li Wang, Teng Zhao, Yongcheng Song, Zhijun Zhang, Huisong Zhang, Jin Liu, Jian Li

Visual relocalization estimates the precise pose of a query image within a pre-built visual map, serving as a fundamental component for robot navigation, autonomous driving, surveying and mapping, etc. In the past few decades, significant research efforts have been devoted to achieving high relocalization accuracy. However, challenges remain when the query images exhibit significant changes compared to the reference scene. This paper primarily addresses the problem of pose verification and correction of inaccurate pose estimations from the relocalization. We propose a novel anchor-based visual relocalization framework that achieves robust pose estimations through multi-view co-visibility verification. Our approach further utilizes a tightly-coupled multi-sensor data fusion for pose refinement. Comprehensive evaluations on large-scale, real-world urban driving datasets (containing frequent dynamic objects, severe occlusions, and long-term environmental changes) demonstrate that our framework achieves state-of-the-art performance. Specifically, compared to traditional SFM-based and Transformer-based methods under these challenging conditions, our approach reduces the translation error by 46.2% and the rotation error by 8.55%.

视觉重定位是在预先构建的视觉地图中估计查询图像的精确姿态，是机器人导航、自动驾驶、测绘等的基本组成部分。在过去的几十年里，大量的研究工作致力于实现高的再定位精度。然而，与参考场景相比，当查询图像显示出显著变化时，挑战仍然存在。本文主要解决了姿态验证问题和姿态估计不准确的修正问题。提出了一种新的基于锚点的视觉定位框架，该框架通过多视图共可视性验证实现鲁棒姿态估计。我们的方法进一步利用紧密耦合的多传感器数据融合来进行姿态优化。对大规模、真实的城市驾驶数据集（包含频繁的动态物体、严重的闭塞和长期的环境变化）的综合评估表明，我们的框架达到了最先进的性能。具体来说，在这些具有挑战性的条件下，与传统的基于sfm和基于transformer的方法相比，我们的方法将平移误差降低了46.2%，旋转误差降低了8.55%。

{"title":"AnchorReF: A novel anchor-based visual re-localization framework aided by multi-sensor data fusion","authors":"Hao Wu , Yu Ran , Xiaoxiang Zhang, Xinying Luo, Li Wang, Teng Zhao, Yongcheng Song, Zhijun Zhang, Huisong Zhang, Jin Liu, Jian Li","doi":"10.1016/j.isprsjprs.2026.01.019","DOIUrl":"10.1016/j.isprsjprs.2026.01.019","url":null,"abstract":"<div><div>Visual relocalization estimates the precise pose of a query image within a pre-built visual map, serving as a fundamental component for robot navigation, autonomous driving, surveying and mapping, etc. In the past few decades, significant research efforts have been devoted to achieving high relocalization accuracy. However, challenges remain when the query images exhibit significant changes compared to the reference scene. This paper primarily addresses the problem of pose verification and correction of inaccurate pose estimations from the relocalization. We propose a novel anchor-based visual relocalization framework that achieves robust pose estimations through multi-view co-visibility verification. Our approach further utilizes a tightly-coupled multi-sensor data fusion for pose refinement. Comprehensive evaluations on large-scale, real-world urban driving datasets (containing frequent dynamic objects, severe occlusions, and long-term environmental changes) demonstrate that our framework achieves state-of-the-art performance. Specifically, compared to traditional SFM-based and Transformer-based methods under these challenging conditions, our approach reduces the translation error by 46.2% and the rotation error by 8.55%.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"233 ","pages":"Pages 1-13"},"PeriodicalIF":12.2,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145982032","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}

引用次数: 0

Varying sensitivities of RED-NIR-based vegetation indices to the input reflectance affect the detected long-term trends 基于red - nir的植被指数对输入反射率的不同敏感性影响了探测到的长期趋势

IF 12.2 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2026-03-01 Epub Date: 2026-01-23 DOI: 10.1016/j.isprsjprs.2026.01.028

Qing Tian , Hongxiao Jin , Rasmus Fensholt , Torbern Tagesson , Luwei Feng , Feng Tian

Widespread vegetation changes have been evidenced by satellite-observed long-term trends over decades in vegetation indices (VIs). However, many issues can affect the derived VIs trends, among which the inherent difference between VIs calculated from the same input reflectance has not been investigated. Here, we compared global long-term trends in six widely used RED-NIR (near-infrared)-based VIs calculated from the MODIS nadir bidirectional reflectance distribution function (BRDF) adjusted product (MCD43A4) during 2000–2023, including normalized difference vegetation index (NDVI), kernel NDVI (kNDVI), 2-band enhanced vegetation index (EVI2), near-infrared reflectance of vegetation (NIRv), difference vegetation index (DVI), and plant phenology index (PPI). We identified two distinct groups of VIs, i.e., (1) NDVI and kNDVI, and (2) EVI2, NIRv, DVI, and PPI, which shared similar trends within the group but showed significant directional differences between groups in 17.4% of the studied area. Only 20.5% of the global land surface showed consistent trends. Based on the radiation transfer model and remote sensing observations, we demonstrated that the two groups of VIs differed in their sensitivities to RED and NIR reflectance. These differences lead to inconsistent long-term trends arising from variations in vegetation type, mixed pixel effects, saturation, and asynchronous changes in vegetation chlorophyll content and structural attributes. Comparisons with ground-observed leaf area index (LAI), flux tower gross primary productivity (GPP), and PhenoCam green chromatic coordinate (GCC) further revealed that the EVI2, NIRv, DVI, and PPI trends corresponded more closely with LAI and GPP trends, whereas the NDVI and kNDVI trends were more strongly associated with GCC trends. Our results highlight that long-term vegetation trends derived from different RED–NIR-based VIs must be interpreted by considering their intrinsic sensitivities to biophysical properties, which is essential for reliable assessments of vegetation dynamics.

卫星观测到的植被指数（VIs）几十年来的长期趋势证明了广泛的植被变化。然而，许多问题会影响所得的能见度趋势，其中，从相同的输入反射率计算所得的能见度之间的固有差异尚未得到研究。本文比较了MODIS最低值双向反射率分布函数（BRDF）调整后的产品（MCD43A4）在2000-2023年间全球广泛使用的6种基于RED-NIR（近红外）的VIs的长期趋势，包括归一化差异植被指数（NDVI）、核NDVI （kNDVI）、2波段增强植被指数（EVI2）、植被近红外反射率（NIRv）、差异植被指数（DVI）和植物物象指数（PPI）。我们确定了两个不同的VIs组，即(1)NDVI和kNDVI， (2) EVI2、NIRv、DVI和PPI，它们在组内具有相似的趋势，但在17.4%的研究区域中，组间存在显著的方向性差异。全球只有20.5%的陆地表面呈现出一致的趋势。基于辐射传输模型和遥感观测，我们证明了两组VIs对红、近红外反射率的敏感性不同。这些差异导致植被类型、混合像元效应、饱和度的变化以及植被叶绿素含量和结构属性的非同步变化导致长期趋势不一致。与地面观测叶面积指数（LAI）、通量塔总初级生产力（GPP）和PhenoCam绿色坐标（GCC）的比较进一步表明，EVI2、NIRv、DVI和PPI趋势与LAI和GPP趋势的相关性更强，而NDVI和kNDVI趋势与GCC趋势的相关性更强。我们的研究结果强调，从不同的基于red - nir的VIs中得出的长期植被趋势必须考虑到它们对生物物理特性的内在敏感性，这对于可靠地评估植被动态至关重要。

{"title":"Varying sensitivities of RED-NIR-based vegetation indices to the input reflectance affect the detected long-term trends","authors":"Qing Tian , Hongxiao Jin , Rasmus Fensholt , Torbern Tagesson , Luwei Feng , Feng Tian","doi":"10.1016/j.isprsjprs.2026.01.028","DOIUrl":"10.1016/j.isprsjprs.2026.01.028","url":null,"abstract":"<div><div>Widespread vegetation changes have been evidenced by satellite-observed long-term trends over decades in vegetation indices (VIs). However, many issues can affect the derived VIs trends, among which the inherent difference between VIs calculated from the same input reflectance has not been investigated. Here, we compared global long-term trends in six widely used RED-NIR (near-infrared)-based VIs calculated from the MODIS nadir bidirectional reflectance distribution function (BRDF) adjusted product (MCD43A4) during 2000–2023, including normalized difference vegetation index (NDVI), kernel NDVI (kNDVI), 2-band enhanced vegetation index (EVI2), near-infrared reflectance of vegetation (NIRv), difference vegetation index (DVI), and plant phenology index (PPI). We identified two distinct groups of VIs, i.e., (1) NDVI and kNDVI, and (2) EVI2, NIRv, DVI, and PPI, which shared similar trends within the group but showed significant directional differences between groups in 17.4% of the studied area. Only 20.5% of the global land surface showed consistent trends. Based on the radiation transfer model and remote sensing observations, we demonstrated that the two groups of VIs differed in their sensitivities to RED and NIR reflectance. These differences lead to inconsistent long-term trends arising from variations in vegetation type, mixed pixel effects, saturation, and asynchronous changes in vegetation chlorophyll content and structural attributes. Comparisons with ground-observed leaf area index (LAI), flux tower gross primary productivity (GPP), and PhenoCam green chromatic coordinate (GCC) further revealed that the EVI2, NIRv, DVI, and PPI trends corresponded more closely with LAI and GPP trends, whereas the NDVI and kNDVI trends were more strongly associated with GCC trends. Our results highlight that long-term vegetation trends derived from different RED–NIR-based VIs must be interpreted by considering their intrinsic sensitivities to biophysical properties, which is essential for reliable assessments of vegetation dynamics.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"233 ","pages":"Pages 247-265"},"PeriodicalIF":12.2,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146033656","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}

引用次数: 0

Comparative assessment of AI-based and classical DSAS approaches in multi-temporal shoreline prediction: A case study of Ras El-Bar coast, Egypt 基于人工智能和经典DSAS方法在多时间线预测中的比较评估：以埃及Ras El-Bar海岸为例

IF 12.2 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2026-03-01 Epub Date: 2026-02-01 DOI: 10.1016/j.isprsjprs.2026.01.040

Hesham M. El-Asmar, Mahmoud Sh. Felfla

Along Ras El-Bar coast, NE Nile Delta of Egypt, intense human interventions and natural processes give rise to highly distinctive non-linear shoreline dynamics that challenge the stationarity assumptions of traditional forecasting. Classical methods, specifically the Digital Shoreline Analysis System Linear Regression Rate (DSAS-LRR), often fail to capture abrupt anthropogenic regime shifts induced by engineering structures. This study presents a comparative assessment of DSAS-LRR against two artificial intelligence (AI) recurrent neural networks, Long Short-Term Memory (LSTM) and Nonlinear Autoregressive Exogenous (NARX), using a multi-decadal satellite-derived shoreline record (1982–2024) to project shoreline evolution through 2050. Results show DSAS-LRR unrealistic projections, exceeding 250 m of displacement by 2050, due to its inability to account for rapid anthropogenic interventions. Conversely, AI models successfully captured complex temporal responses. While LSTM provided conservative estimates, the intervention-aware NARX model achieved the highest predictive accuracy and spatial consistency. Model performance was rigorously evaluated via Taylor diagrams, Performance Index Metric (PIm), and transect-based RMSE analysis, and was further validated by an independent 2025 “blind test”. NARX consistently outperformed both models, accurately reproducing accretion in breakwater shadow zones and moderate erosion between structures, with average RMSE values of 6–14 m. These findings underscore that for anthropogenically modified coasts, intervention-aware AI is no longer just an alternative, it is an essential tool for reliable prediction. The proposed framework provides a transferable roadmap for evidence-based coastal management and infrastructure planning in vulnerable deltaic systems worldwide.

在埃及新尼罗河三角洲Ras El-Bar海岸，强烈的人类干预和自然过程产生了高度独特的非线性海岸线动态，挑战了传统预测的平稳性假设。传统的方法，特别是数字海岸线分析系统线性回归率（DSAS-LRR），往往无法捕捉到由工程结构引起的人为突变。本研究利用多年卫星岸线记录（1982-2024）预测到2050年的海岸线演变，对DSAS-LRR与两种人工智能（AI）递归神经网络，长短期记忆（LSTM）和非线性自回归外生（NARX）进行了比较评估。结果表明，由于无法考虑快速的人为干预，DSAS-LRR的预测不现实，到2050年将超过250米的位移。相反，人工智能模型成功捕获了复杂的时间反应。LSTM提供了保守估计，而干预感知的NARX模型具有最高的预测精度和空间一致性。模型的性能通过泰勒图、性能指标度量（PIm）和基于横断面的RMSE分析进行严格评估，并通过独立的2025“盲测”进一步验证。NARX一直优于这两种模型，准确地再现了防波堤阴影区的增生和结构之间的适度侵蚀，平均RMSE值为6-14 m。这些发现强调，对于人为改变的海岸，具有干预意识的人工智能不再仅仅是一种选择，它是可靠预测的重要工具。提出的框架为全球脆弱的三角洲系统的沿海管理和基础设施规划提供了一个可转移的路线图。

{"title":"Comparative assessment of AI-based and classical DSAS approaches in multi-temporal shoreline prediction: A case study of Ras El-Bar coast, Egypt","authors":"Hesham M. El-Asmar, Mahmoud Sh. Felfla","doi":"10.1016/j.isprsjprs.2026.01.040","DOIUrl":"10.1016/j.isprsjprs.2026.01.040","url":null,"abstract":"<div><div>Along Ras El-Bar coast, NE Nile Delta of Egypt, intense human interventions and natural processes give rise to highly distinctive non-linear shoreline dynamics that challenge the stationarity assumptions of traditional forecasting. Classical methods, specifically the Digital Shoreline Analysis System Linear Regression Rate (DSAS-LRR), often fail to capture abrupt anthropogenic regime shifts induced by engineering structures. This study presents a comparative assessment of DSAS-LRR against two artificial intelligence (AI) recurrent neural networks, Long Short-Term Memory (LSTM) and Nonlinear Autoregressive Exogenous (NARX), using a multi-decadal satellite-derived shoreline record (1982–2024) to project shoreline evolution through 2050. Results show DSAS-LRR unrealistic projections, exceeding 250 m of displacement by 2050, due to its inability to account for rapid anthropogenic interventions. Conversely, AI models successfully captured complex temporal responses. While LSTM provided conservative estimates, the intervention-aware NARX model achieved the highest predictive accuracy and spatial consistency. Model performance was rigorously evaluated via Taylor diagrams, Performance Index Metric (PIm), and transect-based RMSE analysis, and was further validated by an independent 2025 “blind test”. NARX consistently outperformed both models, accurately reproducing accretion in breakwater shadow zones and moderate erosion between structures, with average RMSE values of 6–14 m. These findings underscore that for anthropogenically modified coasts, intervention-aware AI is no longer just an alternative, it is an essential tool for reliable prediction. The proposed framework provides a transferable roadmap for evidence-based coastal management and infrastructure planning in vulnerable deltaic systems worldwide.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"233 ","pages":"Pages 407-422"},"PeriodicalIF":12.2,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146095922","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}

引用次数: 0

L2M-Reg: Building-level uncertainty-aware registration of outdoor LiDAR point clouds and semantic 3D city models L2M-Reg：室外激光雷达点云和语义三维城市模型的建筑级不确定性感知配准

IF 12.2 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2026-03-01 Epub Date: 2026-02-06 DOI: 10.1016/j.isprsjprs.2026.02.005

Ziyang Xu , Benedikt Schwab , Yihui Yang , Thomas H. Kolbe , Christoph Holst

Accurate registration between LiDAR (Light Detection and Ranging) point clouds and semantic 3D city models is a fundamental topic in urban digital twinning and a prerequisite for downstream tasks, such as digital construction, change detection, and model refinement. However, achieving accurate LiDAR-to-Model registration at the individual building level remains challenging, particularly due to the generalization uncertainty in semantic 3D city models at the Level of Detail 2 (LoD2). This paper addresses this gap by proposing L2M-Reg, a plane-based fine registration method that explicitly accounts for model uncertainty. L2M-Reg consists of three key steps: establishing reliable plane correspondence, building a pseudo-plane-constrained Gauss–Helmert model, and adaptively estimating vertical translation. Overall, extensive experiments on five real-world datasets demonstrate that L2M-Reg is both more accurate and computationally efficient than current leading ICP-based and plane-based methods. Therefore, L2M-Reg provides a novel building-level solution regarding LiDAR-to-Model registration when model uncertainty is present. The datasets and code for L2M-Reg can be found: https://github.com/Ziyang-Geodesy/L2M-Reg.

激光雷达（光探测和测距）点云和语义三维城市模型之间的准确配准是城市数字孪生的基本课题，也是数字建设、变化检测和模型细化等下游任务的先决条件。然而，在单个建筑层面实现准确的激光雷达到模型注册仍然具有挑战性，特别是由于语义3D城市模型在细节2级（LoD2）的泛化不确定性。本文通过提出L2M-Reg解决了这一差距，这是一种基于平面的精细配准方法，明确地考虑了模型的不确定性。L2M-Reg包括三个关键步骤：建立可靠的平面对应关系，建立伪平面约束的Gauss-Helmert模型，以及自适应估计垂直平移。总体而言，在五个真实世界数据集上进行的大量实验表明，L2M-Reg比目前领先的基于icp和基于平面的方法更准确，计算效率更高。因此，当存在模型不确定性时，L2M-Reg为激光雷达到模型的注册提供了一种新颖的建筑级解决方案。L2M-Reg的数据集和代码可以在https://github.com/Ziyang-Geodesy/L2M-Reg找到。

{"title":"L2M-Reg: Building-level uncertainty-aware registration of outdoor LiDAR point clouds and semantic 3D city models","authors":"Ziyang Xu , Benedikt Schwab , Yihui Yang , Thomas H. Kolbe , Christoph Holst","doi":"10.1016/j.isprsjprs.2026.02.005","DOIUrl":"10.1016/j.isprsjprs.2026.02.005","url":null,"abstract":"<div><div>Accurate registration between LiDAR (Light Detection and Ranging) point clouds and semantic 3D city models is a fundamental topic in urban digital twinning and a prerequisite for downstream tasks, such as digital construction, change detection, and model refinement. However, achieving accurate LiDAR-to-Model registration at the individual building level remains challenging, particularly due to the generalization uncertainty in semantic 3D city models at the Level of Detail 2 (LoD2). This paper addresses this gap by proposing L2M-Reg, a plane-based fine registration method that explicitly accounts for model uncertainty. L2M-Reg consists of three key steps: establishing reliable plane correspondence, building a pseudo-plane-constrained Gauss–Helmert model, and adaptively estimating vertical translation. Overall, extensive experiments on five real-world datasets demonstrate that L2M-Reg is both more accurate and computationally efficient than current leading ICP-based and plane-based methods. Therefore, L2M-Reg provides a novel building-level solution regarding LiDAR-to-Model registration when model uncertainty is present. The datasets and code for L2M-Reg can be found: <span><span>https://github.com/Ziyang-Geodesy/L2M-Reg</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"233 ","pages":"Pages 467-486"},"PeriodicalIF":12.2,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146135038","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}

引用次数: 0

A near-real-time multi-temporal polarimetric InSAR method for landslides monitoring in rapid-decorrelation scenarios 快速去相关情景下近实时多时相极化InSAR滑坡监测方法

IF 12.2 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2026-03-01 Epub Date: 2026-02-09 DOI: 10.1016/j.isprsjprs.2026.02.006

Yaogang Chen , Jun Hu , Jordi J. Mallorqui , Haiqiang Fu , Wanji Zheng , Aoqing Guo

Interferometric synthetic aperture radar (InSAR) technology can measure ground deformation with high precision over wide areas, which is essential for understanding natural hazards and ensuring infrastructure safety. However, in regions with dense vegetation or frequent surface changes, the radar echoes lose stability over time due to temporal decorrelation. This severely limits the reliability and accuracy of InSAR measurements. Many advanced processing methods have been developed to address this issue, and while they work well in stable conditions, their performance degrades sharply when coherence is lost rapidly. To overcome this limitation, this study proposes a near-real-time sequential multi-temporal polarimetric InSAR (MT-PolInSAR) method tailored for such conditions. For each new acquisition, a stack comprising only the latest images is formed, and statistically homogeneous pixels are reselected dynamically to adapt to evolving scattering mechanisms. A sequential polarimetric-temporal phase optimization is then applied within the stack that confines estimation to short, high-coherence windows and avoids coherence loss between stacks, thereby reducing the effect of fast temporal decorrelation. Deformation time series are subsequently updated through a sequential least squares (LS) inversion using only the newly formed interferograms, which eliminates the need to reprocess the whole dataset and enables timely updates. Experiments with simulated data and full-polarization ALOS-2 and dual-polarization Sentinel-1 images over Fengjie, China, demonstrate that the proposed method significantly increases coherent pixel density and improves deformation accuracy in rapid-decorrelation areas, while enabling genuine near-real-time monitoring with a more efficient processing strategy.

干涉合成孔径雷达（InSAR）技术可以在大范围内高精度测量地面变形，这对于了解自然灾害和确保基础设施安全至关重要。然而，在植被密集或地表变化频繁的地区，雷达回波由于时间去相关而失去稳定性。这严重限制了InSAR测量的可靠性和准确性。许多先进的处理方法已经开发出来解决这个问题，虽然它们在稳定条件下工作良好，但当相干性迅速丧失时，它们的性能急剧下降。为了克服这一限制，本研究提出了一种针对这种情况量身定制的近实时序列多时相偏振InSAR （MT-PolInSAR）方法。对于每一个新的采集，只形成一个由最新图像组成的堆栈，并动态地重新选择统计上均匀的像素以适应不断变化的散射机制。然后在堆栈内应用顺序极化-时间相位优化，将估计限制在短的高相干窗口，避免堆栈之间的相干损失，从而减少快速时间去相关的影响。变形时间序列随后仅使用新形成的干涉图通过顺序最小二乘（LS）反演进行更新，从而消除了对整个数据集进行重新处理的需要，并实现了及时更新。利用模拟数据和全偏振ALOS-2和双偏振Sentinel-1图像在中国Fengjie上空进行的实验表明，该方法显著提高了快速去相关区域的相干像元密度和变形精度，同时以更高效的处理策略实现了真正的近实时监测。

{"title":"A near-real-time multi-temporal polarimetric InSAR method for landslides monitoring in rapid-decorrelation scenarios","authors":"Yaogang Chen , Jun Hu , Jordi J. Mallorqui , Haiqiang Fu , Wanji Zheng , Aoqing Guo","doi":"10.1016/j.isprsjprs.2026.02.006","DOIUrl":"10.1016/j.isprsjprs.2026.02.006","url":null,"abstract":"<div><div>Interferometric synthetic aperture radar (InSAR) technology can measure ground deformation with high precision over wide areas, which is essential for understanding natural hazards and ensuring infrastructure safety. However, in regions with dense vegetation or frequent surface changes, the radar echoes lose stability over time due to temporal decorrelation. This severely limits the reliability and accuracy of InSAR measurements. Many advanced processing methods have been developed to address this issue, and while they work well in stable conditions, their performance degrades sharply when coherence is lost rapidly. To overcome this limitation, this study proposes a near-real-time sequential multi-temporal polarimetric InSAR (MT-PolInSAR) method tailored for such conditions. For each new acquisition, a stack comprising only the latest images is formed, and statistically homogeneous pixels are reselected dynamically to adapt to evolving scattering mechanisms. A sequential polarimetric-temporal phase optimization is then applied within the stack that confines estimation to short, high-coherence windows and avoids coherence loss between stacks, thereby reducing the effect of fast temporal decorrelation. Deformation time series are subsequently updated through a sequential least squares (LS) inversion using only the newly formed interferograms, which eliminates the need to reprocess the whole dataset and enables timely updates. Experiments with simulated data and full-polarization ALOS-2 and dual-polarization Sentinel-1 images over Fengjie, China, demonstrate that the proposed method significantly increases coherent pixel density and improves deformation accuracy in rapid-decorrelation areas, while enabling genuine near-real-time monitoring with a more efficient processing strategy.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"233 ","pages":"Pages 556-574"},"PeriodicalIF":12.2,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146146858","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}

引用次数: 0

Minimizing sun glint impacts on Sentinel-2 Sargassum mapping 最小化太阳闪烁对Sentinel-2马尾藻绘图的影响

IF 12.2 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2026-03-01 Epub Date: 2026-02-12 DOI: 10.1016/j.isprsjprs.2026.02.012

Junnan Jiao, Chuanmin Hu, Brian B. Barnes

Sargassum inundations around the Caribbean Sea and Gulf of Mexico, especially since the emergence of the Great Atlantic Sargassum Belt in 2011, have promoted the use of various remote sensing techniques in monitoring and tracking this brown macroalgae. Among these are the Sentinel-2 MultiSpectral Instruments (MSI) that provide 10-m ground resolution and 5-day revisits for subtropical and tropical waters. However, MSI imagery is often impacted by strong sun glint during the Sargassum season of March–September, creating data gaps or inconsistencies in the estimated Sargassum density among adjacent paths. Here, using overlapping pixels between Sentinel-2A and Sentinel-2B images where only one of them is under strong sun glint, we develop an empirical correction scheme to force Sargassum estimates under strong sun glint to agree with those under minimal sun glint. The correction is based on the surface roughness estimated directly from statistics of Sargassum-free water pixels near the Sargassum patches. Application of the correction scheme shows that the glint-induced overestimation of Sargassum density has been reduced substantially, with mean absolute error (MAE) decreased from 22.8% to 6.3% and root mean square error (RMSE) dropped from 25.6% to 7.7%. These results demonstrate that the proposed approach can substantially reduce uncertainties in Sargassum quantification from both Sentinel-2A and Sentinel-2B and therefore support its integration into the currently operational Sargassum Watch System (SaWS).

加勒比海和墨西哥湾周围的马尾藻泛滥，特别是自2011年大西洋马尾藻带出现以来，促进了各种遥感技术在监测和跟踪这种棕色大型藻类方面的应用。其中包括Sentinel-2多光谱仪器（MSI），它提供10米的地面分辨率和对亚热带和热带水域的5天回访。然而，在3月至9月的马尾藻季节，MSI图像经常受到强烈太阳闪光的影响，导致相邻路径上马尾藻密度的估计出现数据缺口或不一致。在这里，我们利用Sentinel-2A和Sentinel-2B图像之间的重叠像素，其中只有一幅图像处于强烈的太阳闪烁下，我们开发了一种经验校正方案，以迫使强烈太阳闪烁下的马尾藻估计与最小太阳闪烁下的马尾藻估计一致。校正是基于马尾藻斑块附近无马尾藻水像元的统计数据直接估计的表面粗糙度。校正方案的应用表明，闪烁引起的马尾藻密度高估得到了显著降低，平均绝对误差（MAE）从22.8%下降到6.3%，均方根误差（RMSE）从25.6%下降到7.7%。这些结果表明，所提出的方法可以大大减少Sentinel-2A和Sentinel-2B中马尾藻量化的不确定性，因此支持将其集成到当前运行的马尾藻观察系统（SaWS）中。

{"title":"Minimizing sun glint impacts on Sentinel-2 Sargassum mapping","authors":"Junnan Jiao, Chuanmin Hu, Brian B. Barnes","doi":"10.1016/j.isprsjprs.2026.02.012","DOIUrl":"10.1016/j.isprsjprs.2026.02.012","url":null,"abstract":"<div><div><em>Sargassum</em> inundations around the Caribbean Sea and Gulf of Mexico, especially since the emergence of the Great Atlantic <em>Sargassum</em> Belt in 2011, have promoted the use of various remote sensing techniques in monitoring and tracking this brown macroalgae. Among these are the Sentinel-2 MultiSpectral Instruments (MSI) that provide 10-m ground resolution and 5-day revisits for subtropical and tropical waters. However, MSI imagery is often impacted by strong sun glint during the <em>Sargassum</em> season of March–September, creating data gaps or inconsistencies in the estimated <em>Sargassum</em> density among adjacent paths. Here, using overlapping pixels between Sentinel-2A and Sentinel-2B images where only one of them is under strong sun glint, we develop an empirical correction scheme to force <em>Sargassum</em> estimates under strong sun glint to agree with those under minimal sun glint. The correction is based on the surface roughness estimated directly from statistics of <em>Sargassum</em>-free water pixels near the <em>Sargassum</em> patches. Application of the correction scheme shows that the glint-induced overestimation of <em>Sargassum</em> density has been reduced substantially, with mean absolute error (MAE) decreased from 22.8% to 6.3% and root mean square error (RMSE) dropped from 25.6% to 7.7%. These results demonstrate that the proposed approach can substantially reduce uncertainties in <em>Sargassum</em> quantification from both Sentinel-2A and Sentinel-2B and therefore support its integration into the currently operational <em>Sargassum</em> Watch System (SaWS).</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"233 ","pages":"Pages 651-663"},"PeriodicalIF":12.2,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146209466","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}

引用次数: 0

Set-CVGL: A new perspective on cross-view geo-localization with unordered ground-view image sets Set-CVGL：一种基于无序地视图像集的跨视点地理定位新视角

IF 12.2 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2026-03-01 Epub Date: 2026-01-30 DOI: 10.1016/j.isprsjprs.2026.01.037

Qiong Wu , Panwang Xia , Lei Yu , Yi Liu , Mingtao Xiong , Liheng Zhong , Jingdong Chen , Ming Yang , Yongjun Zhang , Yi Wan

Cross-view geo-localization (CVGL) has been widely applied in fields such as robotic navigation and geographic information coupling. Existing approaches primarily use single images or fixed-view image sequences as queries, which limits perspective diversity. In contrast, when humans determine their location visually, they typically move around to gather multiple perspectives. This behavior suggests that integrating diverse visual cues can improve geo-localization reliability. Therefore, we propose a novel task: Cross-View Image Set Geo-Localization (Set-CVGL), which gathers multiple images with diverse perspectives as a query set for localization. To support this task, we introduce SetVL-480K, a benchmark comprising 480,000 ground images captured worldwide and their corresponding satellite images, with each satellite image corresponds to an average of 40 ground images from varied perspectives and locations. Furthermore, we propose FlexGeo, a flexible method designed for Set-CVGL that can also adapt to single-image and image-sequence inputs. FlexGeo includes two key modules: the Similarity-guided Feature Fuser (SFF), which adaptively fuses image features without prior content dependency, and the Individual-level Attributes Learner (IAL), leveraging geo-attributes of each image for comprehensive scene perception. FlexGeo consistently outperforms existing methods on SetVL-480K and four public datasets (VIGOR, University-1652, SeqGeo, and KITTI-CVL), achieving a 2.34

\times

improvement in localization accuracy on SetVL-480K. The codes and dataset will be available at https://github.com/Mabel0403/Set-CVGL.

交叉视角地理定位技术在机器人导航、地理信息耦合等领域有着广泛的应用。现有的方法主要使用单个图像或固定视图图像序列作为查询，这限制了视角的多样性。相比之下，当人类在视觉上确定自己的位置时，他们通常会四处走动，以收集多个视角。这种行为表明，整合不同的视觉线索可以提高地理定位的可靠性。因此，我们提出了一种新的任务：交叉视图图像集地理定位（Set- cvgl），该任务将具有不同视角的多幅图像作为定位的查询集。为了支持这项任务，我们引入了SetVL-480K，这是一个基准，包括48万张全球地面图像及其相应的卫星图像，每张卫星图像平均对应40张不同角度和位置的地面图像。此外，我们提出了FlexGeo，这是一种为Set-CVGL设计的灵活方法，也可以适应单图像和图像序列输入。FlexGeo包括两个关键模块：相似性引导的特征融合器（SFF），它自适应地融合图像特征，而不依赖于先前的内容；以及个人层面的属性学习器（IAL），利用每张图像的地理属性进行全面的场景感知。FlexGeo在SetVL-480K和四个公共数据集（VIGOR、University-1652、SeqGeo和KITTI-CVL）上的定位精度持续优于现有方法，在SetVL-480K上的定位精度提高了2.34倍。代码和数据集可在https://github.com/Mabel0403/Set-CVGL上获得。

{"title":"Set-CVGL: A new perspective on cross-view geo-localization with unordered ground-view image sets","authors":"Qiong Wu , Panwang Xia , Lei Yu , Yi Liu , Mingtao Xiong , Liheng Zhong , Jingdong Chen , Ming Yang , Yongjun Zhang , Yi Wan","doi":"10.1016/j.isprsjprs.2026.01.037","DOIUrl":"10.1016/j.isprsjprs.2026.01.037","url":null,"abstract":"<div><div>Cross-view geo-localization (CVGL) has been widely applied in fields such as robotic navigation and geographic information coupling. Existing approaches primarily use single images or fixed-view image sequences as queries, which limits perspective diversity. In contrast, when humans determine their location visually, they typically move around to gather multiple perspectives. This behavior suggests that integrating diverse visual cues can improve geo-localization reliability. Therefore, we propose a novel task: Cross-View Image Set Geo-Localization (Set-CVGL), which gathers multiple images with diverse perspectives as <strong>a query set</strong> for localization. To support this task, we introduce SetVL-480K, a benchmark comprising 480,000 ground images captured worldwide and their corresponding satellite images, with each satellite image corresponds to an average of 40 ground images from varied perspectives and locations. Furthermore, we propose FlexGeo, a flexible method designed for Set-CVGL that can also adapt to single-image and image-sequence inputs. FlexGeo includes two key modules: the Similarity-guided Feature Fuser (SFF), which adaptively fuses image features without prior content dependency, and the Individual-level Attributes Learner (IAL), leveraging geo-attributes of each image for comprehensive scene perception. FlexGeo consistently outperforms existing methods on SetVL-480K and four public datasets (VIGOR, University-1652, SeqGeo, and KITTI-CVL), achieving a 2.34<span><math><mo>×</mo></math></span> improvement in localization accuracy on SetVL-480K. The codes and dataset will be available at <span><span>https://github.com/Mabel0403/Set-CVGL</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"233 ","pages":"Pages 328-345"},"PeriodicalIF":12.2,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079952","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}

引用次数: 0

Gait-Aware quadruped 3D mapping in challenging environments with complex terrain 复杂地形环境下的步态感知四足三维映射

IF 12.2 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2026-03-01 Epub Date: 2026-02-12 DOI: 10.1016/j.isprsjprs.2026.02.010

Xinge Zhao , Xing Zhang , Yiqing Ni , Qingquan Li , Wentao Zeng , Haixia Feng , Yibo Zhou

Quadruped SLAM has emerged as a promising solution for 3D mapping in complex environment where conventional unmanned aerial vehicles (UAVs) and ground vehicles (UGVs) suffer from mobility constraints. However, the frequent gait transitions, unstable ground contacts, and abrupt body motions of quadruped robots often introduce severe motion distortions and accumulated drift, significantly undermining the robustness of existing SLAM systems. To overcome these challenges, we propose a novel quadruped 3D mapping method that explicitly incorporates gait-aware constraints and confidence-aware updates. First, a gait-aware ESKF-based leg odometry (GA-ESKF) is designed to decouple locomotion state classification from pose estimation, enforcing inter-leg consistency and stance-phase constraints to enhance stability under irregular motion events such as loss of balance. Second, a sequential update strategy tightly integrates GA-ESKF with LiDAR-Inertial SLAM framework, providing high-frequency drift-resilient priors that mitigate short-term IMU propagation errors and reduce LiDAR distortion. Third, a confidence-aware plane-constrained foot-end update mechanism is introduced to mitigate the elevation drift in state estimation arising from complex terrain and agile locomotion of quadruped robots. To validate the proposed approach, we construct a quadruped SLAM dataset covering diverse environments and locomotion behaviors, including walking, stair climbing, running, and destabilizing events such as falling and recovery. Extensive experiments demonstrate that the proposed method significantly outperforms state-of-the-art methods, achieving up to 86.9% reduction in trajectory RMSE and consistently delivering high-fidelity 3D maps under extreme motion conditions. The dataset will be publicly available at https://github.com/EN3D-Lab/Quadruped-SLAM-dataset.

在传统无人机（uav）和地面车辆（ugv）受到机动性限制的复杂环境中，四足SLAM已经成为一种很有前途的3D测绘解决方案。然而，四足机器人频繁的步态转换、不稳定的地面接触和突然的身体运动往往会导致严重的运动扭曲和累积漂移，严重破坏了现有SLAM系统的鲁棒性。为了克服这些挑战，我们提出了一种新的四足3D映射方法，该方法明确地结合了步态感知约束和置信度感知更新。首先，设计了一种基于步态感知eskf的腿部里程计（GA-ESKF），将运动状态分类与姿态估计解耦，加强腿间一致性和姿态相位约束，以增强在不规则运动事件（如失去平衡）下的稳定性。其次，时序更新策略将GA-ESKF与LiDAR-惯性SLAM框架紧密集成，提供高频漂移弹性先验，减轻短期IMU传播误差，减少LiDAR失真。第三，引入置信度感知平面约束下的足端更新机制，缓解四足机器人在复杂地形和敏捷运动条件下状态估计中的高程漂移。为了验证所提出的方法，我们构建了一个四足SLAM数据集，涵盖了不同的环境和运动行为，包括步行、爬楼梯、跑步和不稳定事件（如跌倒和恢复）。大量的实验表明，该方法明显优于最先进的方法，在极端运动条件下，实现了高达86.9%的轨迹RMSE降低，并始终提供高保真的3D地图。该数据集将在https://github.com/EN3D-Lab/Quadruped-SLAM-dataset上公开。

{"title":"Gait-Aware quadruped 3D mapping in challenging environments with complex terrain","authors":"Xinge Zhao , Xing Zhang , Yiqing Ni , Qingquan Li , Wentao Zeng , Haixia Feng , Yibo Zhou","doi":"10.1016/j.isprsjprs.2026.02.010","DOIUrl":"10.1016/j.isprsjprs.2026.02.010","url":null,"abstract":"<div><div>Quadruped SLAM has emerged as a promising solution for 3D mapping in complex environment where conventional unmanned aerial vehicles (UAVs) and ground vehicles (UGVs) suffer from mobility constraints. However, the frequent gait transitions, unstable ground contacts, and abrupt body motions of quadruped robots often introduce severe motion distortions and accumulated drift, significantly undermining the robustness of existing SLAM systems. To overcome these challenges, we propose a novel quadruped 3D mapping method that explicitly incorporates gait-aware constraints and confidence-aware updates. First, a gait-aware ESKF-based leg odometry (GA-ESKF) is designed to decouple locomotion state classification from pose estimation, enforcing inter-leg consistency and stance-phase constraints to enhance stability under irregular motion events such as loss of balance. Second, a sequential update strategy tightly integrates GA-ESKF with LiDAR-Inertial SLAM framework, providing high-frequency drift-resilient priors that mitigate short-term IMU propagation errors and reduce LiDAR distortion. Third, a confidence-aware plane-constrained foot-end update mechanism is introduced to mitigate the elevation drift in state estimation arising from complex terrain and agile locomotion of quadruped robots. To validate the proposed approach, we construct a quadruped SLAM dataset covering diverse environments and locomotion behaviors, including walking, stair climbing, running, and destabilizing events such as falling and recovery. Extensive experiments demonstrate that the proposed method significantly outperforms state-of-the-art methods, achieving up to 86.9% reduction in trajectory RMSE and consistently delivering high-fidelity 3D maps under extreme motion conditions. The dataset will be publicly available at <span><span>https://github.com/EN3D-Lab/Quadruped-SLAM-dataset</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"233 ","pages":"Pages 664-688"},"PeriodicalIF":12.2,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146209467","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}

引用次数: 0