Pub Date : 2026-03-21DOI: 10.1016/j.isprsjprs.2026.03.023
Zhenyu Guo, Kebing Chen, Xuejiao Hou, Lian Feng, Lingling Zhu
{"title":"Suspended sediment decline intensified riverbank erosion and collapse in the Yangtze river mainstream","authors":"Zhenyu Guo, Kebing Chen, Xuejiao Hou, Lian Feng, Lingling Zhu","doi":"10.1016/j.isprsjprs.2026.03.023","DOIUrl":"https://doi.org/10.1016/j.isprsjprs.2026.03.023","url":null,"abstract":"","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"15 9 1","pages":""},"PeriodicalIF":12.7,"publicationDate":"2026-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147496287","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}
Pub Date : 2026-03-18DOI: 10.1016/j.isprsjprs.2026.03.009
Juan Castorena, E. Louise Loudermilk, Scott Pokswinski, Rodman Linn
{"title":"From canopy to ground via ForestGen3D: Learning cross-domain generation of 3D forest structure from aerial-to-terrestrial LiDAR","authors":"Juan Castorena, E. Louise Loudermilk, Scott Pokswinski, Rodman Linn","doi":"10.1016/j.isprsjprs.2026.03.009","DOIUrl":"https://doi.org/10.1016/j.isprsjprs.2026.03.009","url":null,"abstract":"","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"115 1","pages":""},"PeriodicalIF":12.7,"publicationDate":"2026-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147496296","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}
Pub Date : 2026-03-18DOI: 10.1016/j.isprsjprs.2026.03.003
Zhenqi Zheng, Xiao Sun, Xuan Wang, Qiyu Zhang, Bisheng Yang, You Li
{"title":"Lightweight and adaptive fusion framework using remote sensing derived scene backbone map for positioning enhancement in urban NLOS environments","authors":"Zhenqi Zheng, Xiao Sun, Xuan Wang, Qiyu Zhang, Bisheng Yang, You Li","doi":"10.1016/j.isprsjprs.2026.03.003","DOIUrl":"https://doi.org/10.1016/j.isprsjprs.2026.03.003","url":null,"abstract":"","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"52 1","pages":""},"PeriodicalIF":12.7,"publicationDate":"2026-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147496295","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Angular normalization of solar-induced chlorophyll fluorescence on tree crown scale for red and near-infrared bands based on multi-angle UAV observations","authors":"Zhiqiang Cheng, Jing M. Chen, Chunyu Lai, Hongda Zeng, Guofang Miao, Zhiqun Huang","doi":"10.1016/j.isprsjprs.2026.03.022","DOIUrl":"https://doi.org/10.1016/j.isprsjprs.2026.03.022","url":null,"abstract":"","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"29 1","pages":""},"PeriodicalIF":12.7,"publicationDate":"2026-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147496297","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}
Pub Date : 2026-03-17DOI: 10.1016/j.isprsjprs.2026.03.016
Wenxiao Zhan, Jing Chen
3D building instance change detection enables the monitoring of changes in individual building entities, serving as a critical means for security warning and urban management. Existing methods typically adopt a multi-stage approach based on multiple independent task-specific models to sequentially address instance segmentation and change detection in a decoupled manner. However, these methods not only increase computational complexity but also struggle to leverage the synergistic relationships among multiple tasks to enhance change detection performance. Furthermore, existing change detection datasets lack instance-level annotations, which impedes the development of deep learning-based 3D instance change detection methods. Thereby, we expand upon existing 3D point cloud change detection datasets by introducing new building instance labels, resulting in a publicly available urban 3D Building instance Change Detection dataset, named 3DBiCD. It integrates photogrammetric and LiDAR point clouds to characterize 3D building instance change across Hong Kong, China, and Utrecht, Netherlands, with 10,501 building instances, of which 2044 instances change, exhibiting a 19.5% change ratio. Moreover, we propose a Multi-task Synergistic Enhancement Network, named MSENet, which achieves end-to-end 3D building instance change detection in a single-stage approach and leverages inter-task synergy to enhance change detection performance. Specifically, it integrates two innovative modules, the Instance Cluster Mask (ICM)module and the Instance-oriented Change Refinement (ICR)module, to unify instance segmentation and change detection within a single framework. The ICM module processes change-rich encoded features through a clustering-guided feature fusion approach to output precise instance masks, and the ICR module integrates the instance-specific geometric and semantic cues to refine foreground instance change features. Experiments show that MSENet achieves 83.06% and 92.54% of the mean of IoU over building changes on realistic point clouds, leading to an improvement of 3.97% and 2.87% over the state-of-the-art, demonstrating the efficiency of MSENet. The code and dataset are available at https://github.com/zhanwenxiao/MSENet and https://github.com/zhanwenxiao/3DBiCD.
{"title":"MSENet: multi-task synergistic enhancement network for 3D building instance change detection","authors":"Wenxiao Zhan, Jing Chen","doi":"10.1016/j.isprsjprs.2026.03.016","DOIUrl":"https://doi.org/10.1016/j.isprsjprs.2026.03.016","url":null,"abstract":"3D building instance change detection enables the monitoring of changes in individual building entities, serving as a critical means for security warning and urban management. Existing methods typically adopt a multi-stage approach based on multiple independent task-specific models to sequentially address instance segmentation and change detection in a decoupled manner. However, these methods not only increase computational complexity but also struggle to leverage the synergistic relationships among multiple tasks to enhance change detection performance. Furthermore, existing change detection datasets lack instance-level annotations, which impedes the development of deep learning-based 3D instance change detection methods. Thereby, we expand upon existing 3D point cloud change detection datasets by introducing new building instance labels, resulting in a publicly available urban <ce:bold>3D B</ce:bold>uilding <ce:bold>i</ce:bold>nstance <ce:bold>C</ce:bold>hange <ce:bold>D</ce:bold>etection dataset, named 3DBiCD. It integrates photogrammetric and LiDAR point clouds to characterize 3D building instance change across Hong Kong, China, and Utrecht, Netherlands, with 10,501 building instances, of which 2044 instances change, exhibiting a 19.5% change ratio. Moreover, we propose a <ce:bold>M</ce:bold>ulti-task <ce:bold>S</ce:bold>ynergistic <ce:bold>E</ce:bold>nhancement <ce:bold>Net</ce:bold>work, named MSENet, which achieves end-to-end 3D building instance change detection in a single-stage approach and leverages inter-task synergy to enhance change detection performance. Specifically, it integrates two innovative modules, the Instance Cluster Mask (ICM)<ce:hsp sp=\"0.25\"></ce:hsp>module and the Instance-oriented Change Refinement (ICR)<ce:hsp sp=\"0.25\"></ce:hsp>module, to unify instance segmentation and change detection within a single framework. The ICM module processes change-rich encoded features through a clustering-guided feature fusion approach to output precise instance masks, and the ICR module integrates the instance-specific geometric and semantic cues to refine foreground instance change features. Experiments show that MSENet achieves 83.06% and 92.54% of the mean of IoU over building changes on realistic point clouds, leading to an improvement of 3.97% and 2.87% over the state-of-the-art, demonstrating the efficiency of MSENet. The code and dataset are available at https://github.com/zhanwenxiao/MSENet and https://github.com/zhanwenxiao/3DBiCD.","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"270 1","pages":""},"PeriodicalIF":12.7,"publicationDate":"2026-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147464776","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}
Pub Date : 2026-03-16DOI: 10.1016/j.isprsjprs.2026.03.007
Jesse Nowak, Markus Holopainen, Teja Kattenborn, Samuli Junttila
Climate warming is increasing forest disturbances, with hotter summers and longer droughts causing widespread tree mortality. Yet the timing of these deaths remains unclear. While standing dead trees can be delineated from high‑resolution aerial imagery, their infrequent updates limit temporal analysis. Although satellite images have been used to map dead trees, few studies have estimated when trees died using time series data. We present a method to estimate mortality timing within known mortality areas using PlanetScope (3 m) time series in a boreal forest (Finland) and a temperate forest (Luxembourg). We used summer imagery (June–August) from 2021 to 2022 for Helsinki and 2020–2022 for Luxembourg. The reference dataset contained 468 standing dead trees in Helsinki and 3070 in Luxembourg, mapped from very high-resolution aerial (5–10 cm) and satellite images (50 cm). The Helsinki study area was characterized by scattered tree mortality with isolated and small clusters of dead trees, whereas the Luxembourg study area had more clustered tree mortality and larger clusters of dead trees. We evaluated four vegetation indices: kNDVI, GNDVI, SR 800/550, and PSSRc2, and used the Kernelized Change Point Detection (PELT) algorithm to identify sustained declines in pixel values. kNDVI performed best, detecting spectral change for 80% (2832 of 3538) of the dead trees across both areas, with detection increasing with cluster size. In Helsinki, kNDVI detected 204 dead trees (43.6%), while in Luxembourg, 2628 dead trees (86%). The differences in performance between the study areas were mainly attributed to the spatial distribution of tree mortality (scattered vs. clustered). Detection was most reliable for clusters of ≥ 3 trees, while isolated trees were rarely detected, likely due to mixed pixels (only 19% detected). For timing estimation, the overall RMSE across both areas was 245 days with a mean bias of + 6 days (i.e., six days later than visual confirmation). In Helsinki RMSE = 211, bias = − 103; Luxembourg RMSE = 254, bias = +40. The relatively high RMSE alongside a small overall bias indicates variability among individual timing estimates due to reference‑date lags and mixed pixels. With further improvements to change‑point detection and spectral inputs, PlanetScope time series show initial promise for estimating mortality timing in clusters (≥3 trees) of standing deadwood.
{"title":"Estimating tree mortality timing using PlanetScope satellite image time series","authors":"Jesse Nowak, Markus Holopainen, Teja Kattenborn, Samuli Junttila","doi":"10.1016/j.isprsjprs.2026.03.007","DOIUrl":"https://doi.org/10.1016/j.isprsjprs.2026.03.007","url":null,"abstract":"Climate warming is increasing forest disturbances, with hotter summers and longer droughts causing widespread tree mortality. Yet the timing of these deaths remains unclear. While standing dead trees can be delineated from high‑resolution aerial imagery, their infrequent updates limit temporal analysis. Although satellite images have been used to map dead trees, few studies have estimated when trees died using time series data. We present a method to estimate mortality timing within known mortality areas using PlanetScope (3 m) time series in a boreal forest (Finland) and a temperate forest (Luxembourg). We used summer imagery (June–August) from 2021 to 2022 for Helsinki and 2020–2022 for Luxembourg. The reference dataset contained 468 standing dead trees in Helsinki and 3070 in Luxembourg, mapped from very high-resolution aerial (5–10 cm) and satellite images (50 cm). The Helsinki study area was characterized by scattered tree mortality with isolated and small clusters of dead trees, whereas the Luxembourg study area had more clustered tree mortality and larger clusters of dead trees. We evaluated four vegetation indices: kNDVI, GNDVI, SR 800/550, and PSSRc2, and used the Kernelized Change Point Detection (PELT) algorithm to identify sustained declines in pixel values. kNDVI performed best, detecting spectral change for 80% (2832 of 3538) of the dead trees across both areas, with detection increasing with cluster size. In Helsinki, kNDVI detected 204 dead trees (43.6%), while in Luxembourg, 2628 dead trees (86%). The differences in performance between the study areas were mainly attributed to the spatial distribution of tree mortality (scattered vs. clustered). Detection was most reliable for clusters of ≥ 3 trees, while isolated trees were rarely detected, likely due to mixed pixels (only 19% detected). For timing estimation, the overall RMSE across both areas was 245 days with a mean bias of + 6 days (i.e., six days later than visual confirmation). In Helsinki RMSE = 211, bias = − 103; Luxembourg RMSE = 254, bias = +40. The relatively high RMSE alongside a small overall bias indicates variability among individual timing estimates due to reference‑date lags and mixed pixels. With further improvements to change‑point detection and spectral inputs, PlanetScope time series show initial promise for estimating mortality timing in clusters (≥3 trees) of standing deadwood.","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"27 1","pages":""},"PeriodicalIF":12.7,"publicationDate":"2026-03-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147464777","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：