Pub Date : 2024-07-25DOI: 10.5194/isprs-annals-x-4-w5-2024-349-2024
L. Díaz-Vilariño, J. Balado
{"title":"The 19th 3D GeoInfo Conference: Preface Annals","authors":"L. Díaz-Vilariño, J. Balado","doi":"10.5194/isprs-annals-x-4-w5-2024-349-2024","DOIUrl":"https://doi.org/10.5194/isprs-annals-x-4-w5-2024-349-2024","url":null,"abstract":"<jats:p>\u0000 </jats:p>","PeriodicalId":508124,"journal":{"name":"ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences","volume":"55 19","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141804546","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-10DOI: 10.5194/isprs-annals-x-2-2024-105-2024
Yuci Han, C. Toth, Alper Yilmaz
Abstract. The aim of this work is to develop an approach that enables Unmanned Aerial System (UAS) to efficiently learn to navigate in large-scale urban environments and transfer their acquired expertise to novel environments. To achieve this, we propose a metacurriculum training scheme. First, meta-training allows the agent to learn a master policy to generalize across tasks. The resulting model is then fine-tuned on the downstream tasks. We organize the training curriculum in a hierarchical manner such that the agent is guided from coarse to fine towards the target task. In addition, we introduce Incremental Self-Adaptive Reinforcement learning (ISAR), an algorithm that combines the ideas of incremental learning and meta-reinforcement learning (MRL). In contrast to traditional reinforcement learning (RL), which focuses on acquiring a policy for a specific task, MRL aims to learn a policy with fast transfer ability to novel tasks. However, the MRL training process is time consuming, whereas our proposed ISAR algorithm achieves faster convergence than the conventional MRL algorithm. We evaluate the proposed methodologies in simulated environments and demonstrate that using this training philosophy in conjunction with the ISAR algorithm significantly improves the convergence speed for navigation in large-scale cities and the adaptation proficiency in novel environments. The project page is publicly available at https://superhan2611.github.io/isar_nav/.
{"title":"UAS Visual Navigation in Large and Unseen Environments via a Meta Agent","authors":"Yuci Han, C. Toth, Alper Yilmaz","doi":"10.5194/isprs-annals-x-2-2024-105-2024","DOIUrl":"https://doi.org/10.5194/isprs-annals-x-2-2024-105-2024","url":null,"abstract":"Abstract. The aim of this work is to develop an approach that enables Unmanned Aerial System (UAS) to efficiently learn to navigate in large-scale urban environments and transfer their acquired expertise to novel environments. To achieve this, we propose a metacurriculum training scheme. First, meta-training allows the agent to learn a master policy to generalize across tasks. The resulting model is then fine-tuned on the downstream tasks. We organize the training curriculum in a hierarchical manner such that the agent is guided from coarse to fine towards the target task. In addition, we introduce Incremental Self-Adaptive Reinforcement learning (ISAR), an algorithm that combines the ideas of incremental learning and meta-reinforcement learning (MRL). In contrast to traditional reinforcement learning (RL), which focuses on acquiring a policy for a specific task, MRL aims to learn a policy with fast transfer ability to novel tasks. However, the MRL training process is time consuming, whereas our proposed ISAR algorithm achieves faster convergence than the conventional MRL algorithm. We evaluate the proposed methodologies in simulated environments and demonstrate that using this training philosophy in conjunction with the ISAR algorithm significantly improves the convergence speed for navigation in large-scale cities and the adaptation proficiency in novel environments. The project page is publicly available at https://superhan2611.github.io/isar_nav/.\u0000","PeriodicalId":508124,"journal":{"name":"ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences","volume":"104 33","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141361231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-10DOI: 10.5194/isprs-annals-x-2-2024-137-2024
Liangyun Liu, Xiao Zhang
Abstract. Land cover change information plays an indispensable role in environmental monitoring, climate change research, agricultural planning, urban development, biodiversity conservation, and natural disaster risk assessment. Recently, the free access of Landsat imagery and improvement of computation capacity especially supported by Google Earth Engine platform provides great chance in time-series land-cover change monitoring. We used the stratified land-cover monitoring strategy and time-series Landsat imagery to develop a novel global 30 m land-cover dynamic product with fine classification system from 1985 to 2022 (GLC_FCS30D). Firstly, we used the multitemporal classification to generate the time-series impervious surfaces, wetlands and tidal flat products. Then, we proposed to combine the continuous change detection algorithm and local adaptive updating model to capture the land-cover changes, and to generate a new global 30 m land-cover dynamic product (impervious surfaces, wetlands and tidal flat types were excluded in this step). Next, after overlapping the three multitemporal classification products and the time-series dynamical land-cover dataset, the novel GLC_FCS30D was developed, which contained 35 fine land-cover types. Lastly, using the global 84526 validation points in 2020, the GLC_FCS30D was validated to show the great performance with an overall accuracy of 80.88%, and had obvious advantages over other global land-cover products in diversity of land-cover types and mapping accuracy.
{"title":"Algorithm, Progresses, Datasets and Validation of GLC_FCS30D: the first global 30 m land-cover dynamic product with fine classification system from 1985 to 2022","authors":"Liangyun Liu, Xiao Zhang","doi":"10.5194/isprs-annals-x-2-2024-137-2024","DOIUrl":"https://doi.org/10.5194/isprs-annals-x-2-2024-137-2024","url":null,"abstract":"Abstract. Land cover change information plays an indispensable role in environmental monitoring, climate change research, agricultural planning, urban development, biodiversity conservation, and natural disaster risk assessment. Recently, the free access of Landsat imagery and improvement of computation capacity especially supported by Google Earth Engine platform provides great chance in time-series land-cover change monitoring. We used the stratified land-cover monitoring strategy and time-series Landsat imagery to develop a novel global 30 m land-cover dynamic product with fine classification system from 1985 to 2022 (GLC_FCS30D). Firstly, we used the multitemporal classification to generate the time-series impervious surfaces, wetlands and tidal flat products. Then, we proposed to combine the continuous change detection algorithm and local adaptive updating model to capture the land-cover changes, and to generate a new global 30 m land-cover dynamic product (impervious surfaces, wetlands and tidal flat types were excluded in this step). Next, after overlapping the three multitemporal classification products and the time-series dynamical land-cover dataset, the novel GLC_FCS30D was developed, which contained 35 fine land-cover types. Lastly, using the global 84526 validation points in 2020, the GLC_FCS30D was validated to show the great performance with an overall accuracy of 80.88%, and had obvious advantages over other global land-cover products in diversity of land-cover types and mapping accuracy.\u0000","PeriodicalId":508124,"journal":{"name":"ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences","volume":" 19","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141363523","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-10DOI: 10.5194/isprs-annals-x-2-2024-17-2024
Dimitri Bulatov, D. Frommholz, B. Kottler, Kevin Qui, Eva Strauss
Abstract. This paper showcases an integrated workflow hinged on passive airborne multi-modal sensor data for the simulation of the thermal behavior of built-up areas with a focus on urban heat islands. The geometry of the underlying parametrized model, or digital twin, is derived from high-resolution nadir and oblique RGB, near-infrared and thermal infrared imagery. The captured bitmaps get photogrammetrically processed into comprehensive surface models, terrain, dense 3D point clouds and true-ortho mosaics. Building geometries are reconstructed from the projected point sets with procedures presupposing outlining, analysis of roof and fac¸ade details, triangulation, and texturing mapping. For thermal simulation, the composition of the ground is determined using supervised machine learning based on a modified multi-modal DeepLab v3+ architecture. Vegetation is retrieved as individual trees and larger tree regions to be added to the meshed terrain. Building materials are assigned from the available visual, infrared and surface planarity information as well as publicly available references. With actual weather data, surface temperatures can be calculated for any period of time by evaluating conductive, convective, radiative and emissive energy fluxes for triangular layers congruent to the faces of the modeled scene. Results on a sample dataset of the Moabit district in Berlin, Germany, showed the ability of the simulator to output surface temperatures of relatively large datasets efficiently. Compared to the thermal infrared images, several insufficiencies in terms of data and model caused occasional deviations between measured and simulated temperatures. For some of these shortcomings, improvement suggestions within future work are presented.
{"title":"Using Passive Multi-Modal Sensor Data for Thermal Simulation of Urban Surfaces","authors":"Dimitri Bulatov, D. Frommholz, B. Kottler, Kevin Qui, Eva Strauss","doi":"10.5194/isprs-annals-x-2-2024-17-2024","DOIUrl":"https://doi.org/10.5194/isprs-annals-x-2-2024-17-2024","url":null,"abstract":"Abstract. This paper showcases an integrated workflow hinged on passive airborne multi-modal sensor data for the simulation of the thermal behavior of built-up areas with a focus on urban heat islands. The geometry of the underlying parametrized model, or digital twin, is derived from high-resolution nadir and oblique RGB, near-infrared and thermal infrared imagery. The captured bitmaps get photogrammetrically processed into comprehensive surface models, terrain, dense 3D point clouds and true-ortho mosaics. Building geometries are reconstructed from the projected point sets with procedures presupposing outlining, analysis of roof and fac¸ade details, triangulation, and texturing mapping. For thermal simulation, the composition of the ground is determined using supervised machine learning based on a modified multi-modal DeepLab v3+ architecture. Vegetation is retrieved as individual trees and larger tree regions to be added to the meshed terrain. Building materials are assigned from the available visual, infrared and surface planarity information as well as publicly available references. With actual weather data, surface temperatures can be calculated for any period of time by evaluating conductive, convective, radiative and emissive energy fluxes for triangular layers congruent to the faces of the modeled scene. Results on a sample dataset of the Moabit district in Berlin, Germany, showed the ability of the simulator to output surface temperatures of relatively large datasets efficiently. Compared to the thermal infrared images, several insufficiencies in terms of data and model caused occasional deviations between measured and simulated temperatures. For some of these shortcomings, improvement suggestions within future work are presented.\u0000","PeriodicalId":508124,"journal":{"name":"ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences","volume":"118 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141360895","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-10DOI: 10.5194/isprs-annals-x-2-2024-121-2024
Haruna Kawasaki, Saki Komoriya, Hiroshi Masuda
Abstract. In recent years, there has been a high demand in forestry and forest research for the accurate measurement of tree traits from point clouds captured by terrestrial laser scanners. However, the reliability of the calculated values is not sufficient due to the difficulty of accurate registration of each tree over a large area of forest. To solve this problem, we introduce localized multi-view registration for correcting the registration matrix of each tree stem. In addition, we discuss methods for registering the whole point clouds of a forest by using the registration matrices locally calculated for tree stems. Especially, we discuss a method to align tree stem points that do not have sufficient overlapping points required in registration. The proposed method was applied to actual forest point clouds and diameter at breast height (DBH) was compared to the manually measured DBH. Experimental results showed that the proposed method was effective in reducing registration errors and in calculating tree stem traits with high accuracy.
{"title":"Accurate Calculation of Tree Stem Traits in Forests Using Localized Multi-View Registration","authors":"Haruna Kawasaki, Saki Komoriya, Hiroshi Masuda","doi":"10.5194/isprs-annals-x-2-2024-121-2024","DOIUrl":"https://doi.org/10.5194/isprs-annals-x-2-2024-121-2024","url":null,"abstract":"Abstract. In recent years, there has been a high demand in forestry and forest research for the accurate measurement of tree traits from point clouds captured by terrestrial laser scanners. However, the reliability of the calculated values is not sufficient due to the difficulty of accurate registration of each tree over a large area of forest. To solve this problem, we introduce localized multi-view registration for correcting the registration matrix of each tree stem. In addition, we discuss methods for registering the whole point clouds of a forest by using the registration matrices locally calculated for tree stems. Especially, we discuss a method to align tree stem points that do not have sufficient overlapping points required in registration. The proposed method was applied to actual forest point clouds and diameter at breast height (DBH) was compared to the manually measured DBH. Experimental results showed that the proposed method was effective in reducing registration errors and in calculating tree stem traits with high accuracy.\u0000","PeriodicalId":508124,"journal":{"name":"ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences","volume":" 20","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141365180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-10DOI: 10.5194/isprs-annals-x-2-2024-217-2024
P. Schuegraf, Zhixin Li, Jiaojiao Tian, Jie Shan, K. Bittner
Abstract. Nowadays, deep learning allows to automatically learn features from data. Buildings are one of the most important objects in urban environments. They are used in applications such as inputs to building reconstruction, disaster monitoring, city planing and environment modelling for autonomous driving. However, it is not enough to represent them in raster format, since applications require buildings as polygons. We use an existing, learning based approach to extract building footprints from ortho imagery and digital surface model (DSM) and propose a pipeline for building polygon extraction, which we call primary orientation learning (POL). The first step is to extract initial polygons, that contain a vertex for each pixel in the boundary of the footprint. Afterwards, the two primary orientation angles are regressed continuously. Using these orientation, we insert vertices such that all consecutive edges are perpendicular. To the best of our knowledge, our approach is the first to predict a continuous orientation angle for building boundary regularization. Furthermore, the proposed method is highly efficient with an average processing time of 2.879 ms for a single building.
{"title":"Rectilinear Building Footprint Regularization Using Deep Learning","authors":"P. Schuegraf, Zhixin Li, Jiaojiao Tian, Jie Shan, K. Bittner","doi":"10.5194/isprs-annals-x-2-2024-217-2024","DOIUrl":"https://doi.org/10.5194/isprs-annals-x-2-2024-217-2024","url":null,"abstract":"Abstract. Nowadays, deep learning allows to automatically learn features from data. Buildings are one of the most important objects in urban environments. They are used in applications such as inputs to building reconstruction, disaster monitoring, city planing and environment modelling for autonomous driving. However, it is not enough to represent them in raster format, since applications require buildings as polygons. We use an existing, learning based approach to extract building footprints from ortho imagery and digital surface model (DSM) and propose a pipeline for building polygon extraction, which we call primary orientation learning (POL). The first step is to extract initial polygons, that contain a vertex for each pixel in the boundary of the footprint. Afterwards, the two primary orientation angles are regressed continuously. Using these orientation, we insert vertices such that all consecutive edges are perpendicular. To the best of our knowledge, our approach is the first to predict a continuous orientation angle for building boundary regularization. Furthermore, the proposed method is highly efficient with an average processing time of 2.879 ms for a single building.\u0000","PeriodicalId":508124,"journal":{"name":"ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences","volume":"116 46","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141361102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-10DOI: 10.5194/isprs-annals-x-2-2024-163-2024
S. Nietiedt, P. Helmholz, T. Luhmann
Abstract. Dynamic photogrammetry is an established method for acquiring 3D information of deforming objects or dynamic scenes in various close-range applications. A crucial impact has occlusions caused by object deformations, obstacles or camera movements. Temporal occlusions are highly application-specific and sometimes difficult to predict, resulting in a significant reduction of reconstruction quality or the aborting of image sequence processing. Previous approaches usually model such occlusions as semantic information and consider them using image masks. However, generating these image masks requires complex methods and extensive training data. Due to the unpredictability of the complexity and movements of dynamic scenes, generating training data is challenging in many applications. Therefore, this paper proposes an alternative modelling approach, which can be part of a spatio-temporal matching process. Based on the characteristic high redundancy, occlusions can be detected using robust estimation methods and considered in the optimisation. Therefore, no information about the occlusions and further processing steps are necessary. We evaluate our approach with synthetic and real data of an industrial application regarding the accuracy and ability to detect occlusion simultaneously. The evaluation of the proposed approach shows that the impact of occlusion can be eliminated, and the quality of the results is comparable to conventional methods.
{"title":"Occlusion handling in spatio-temporal object-based image sequence matching","authors":"S. Nietiedt, P. Helmholz, T. Luhmann","doi":"10.5194/isprs-annals-x-2-2024-163-2024","DOIUrl":"https://doi.org/10.5194/isprs-annals-x-2-2024-163-2024","url":null,"abstract":"Abstract. Dynamic photogrammetry is an established method for acquiring 3D information of deforming objects or dynamic scenes in various close-range applications. A crucial impact has occlusions caused by object deformations, obstacles or camera movements. Temporal occlusions are highly application-specific and sometimes difficult to predict, resulting in a significant reduction of reconstruction quality or the aborting of image sequence processing. Previous approaches usually model such occlusions as semantic information and consider them using image masks. However, generating these image masks requires complex methods and extensive training data. Due to the unpredictability of the complexity and movements of dynamic scenes, generating training data is challenging in many applications. Therefore, this paper proposes an alternative modelling approach, which can be part of a spatio-temporal matching process. Based on the characteristic high redundancy, occlusions can be detected using robust estimation methods and considered in the optimisation. Therefore, no information about the occlusions and further processing steps are necessary. We evaluate our approach with synthetic and real data of an industrial application regarding the accuracy and ability to detect occlusion simultaneously. The evaluation of the proposed approach shows that the impact of occlusion can be eliminated, and the quality of the results is comparable to conventional methods.\u0000","PeriodicalId":508124,"journal":{"name":"ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences","volume":"108 44","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141362351","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-10DOI: 10.5194/isprs-annals-x-2-2024-65-2024
Ian Estacio, Cristian Román-Palacios, Joseph Hoover, Xiaojiang Li, Chris Lim
Abstract. Urban Green Space (UGS) is vital for improving the public health and sustainability of cities. Vector data on UGS such as open data from governments and OpenStreetMap are available for retrieval by interested users, but the availability of UGS data is still limited on global and temporal scales. This study develops the UGS Extractor, a web-based application for the automatic extraction of UGS given user inputs of Area of Interest and Date of Interest. To accommodate various types of green spaces, such as parks or lawns, the application additionally allows users to set parameters for the minimum size of each UGS and the Minimum Urban Neighbor Density, enabling customization of what qualifies as UGS. The UGS Extractor implements a methodological framework that applies object-based image processing, edge detection and extraction, and image neighborhood analysis on the near real-time 10m Dynamic World collection of Land Use/Land Cover images. The application’s utility was demonstrated through two case studies. In the first, the UGS Extractor accurately mapped major parks when compared to open data sources in New Orleans, USA. In the second, the UGS Extractor demonstrated significant increases in the total area of UGS from 2015 to 2023 in Songdo, South Korea, which consequently improved green space accessibility. These results underscore the UGS Extractor’s utility in extracting specific types of UGS and analyzing their temporal trends. This user-friendly application overall offers higher spatial resolution compared to publicly available satellite-based methods while facilitating temporal studies not possible with vector datasets.
{"title":"Open-source automatic extraction of Urban Green Space: Application to assessing improvement in green space access","authors":"Ian Estacio, Cristian Román-Palacios, Joseph Hoover, Xiaojiang Li, Chris Lim","doi":"10.5194/isprs-annals-x-2-2024-65-2024","DOIUrl":"https://doi.org/10.5194/isprs-annals-x-2-2024-65-2024","url":null,"abstract":"Abstract. Urban Green Space (UGS) is vital for improving the public health and sustainability of cities. Vector data on UGS such as open data from governments and OpenStreetMap are available for retrieval by interested users, but the availability of UGS data is still limited on global and temporal scales. This study develops the UGS Extractor, a web-based application for the automatic extraction of UGS given user inputs of Area of Interest and Date of Interest. To accommodate various types of green spaces, such as parks or lawns, the application additionally allows users to set parameters for the minimum size of each UGS and the Minimum Urban Neighbor Density, enabling customization of what qualifies as UGS. The UGS Extractor implements a methodological framework that applies object-based image processing, edge detection and extraction, and image neighborhood analysis on the near real-time 10m Dynamic World collection of Land Use/Land Cover images. The application’s utility was demonstrated through two case studies. In the first, the UGS Extractor accurately mapped major parks when compared to open data sources in New Orleans, USA. In the second, the UGS Extractor demonstrated significant increases in the total area of UGS from 2015 to 2023 in Songdo, South Korea, which consequently improved green space accessibility. These results underscore the UGS Extractor’s utility in extracting specific types of UGS and analyzing their temporal trends. This user-friendly application overall offers higher spatial resolution compared to publicly available satellite-based methods while facilitating temporal studies not possible with vector datasets.\u0000","PeriodicalId":508124,"journal":{"name":"ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences","volume":"122 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141362714","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-10DOI: 10.5194/isprs-annals-x-2-2024-113-2024
S. Isfort, M. Elias, Hans-Gerd Maas
Abstract. Robust and automated point cloud registration methods are required in many geoscience applications using multi-temporal and multi-modal 3D point clouds. Therefore, a 3D keypoint-based coarse registration workflow has been implemented, utilizing the ISS keypoint detector and 3DSmoothNet descriptor. This paper contributes to keypoint-based registration research through variations of the standard workflow proposed in the literature, applying a two-staged strategy of global and local keypoint matching as well as prototypical keypoint projection and fine registration based on ICP. Further, by testing the utilized detector and descriptor on unstructured, multi-temporal and multi-source point clouds with variations in point cloud density, generalization ability is tested outside benchmark data. Therefore, data of the Bøverbreen glacier in Jotunheimen, Norway has been acquired in 2022 and 2023, deploying UAV-based image matching and terrestrial laser scanning. The results show good performance of the implemented robust matching algorithm PROSAC, requiring fewer iterations than the well-known RANSAC approach, but solving the rigid body transformation with TEASER++ is faster and more robust to outliers without demanding pre-knowledge of the data. Further, the results identify the keypoint detection as most limiting factor in speed and accuracy. Summarizing, keypoint-based coarse registration on low density point clouds, applying a global and local matching strategy and transformation estimation using TEASER++ is recommended. Keypoint projection shows potential, increasing number and precision in low density clouds, but has to be more robust. Further research needs to be carried out, focusing on identifying a fast and robust keypoint detector.
{"title":"Development and Evaluation of a Two-Staged 3D Keypoint Based Workflow for the Co-Registration of Unstructured Multi-Temporal and Multi-Modal 3D Point Clouds","authors":"S. Isfort, M. Elias, Hans-Gerd Maas","doi":"10.5194/isprs-annals-x-2-2024-113-2024","DOIUrl":"https://doi.org/10.5194/isprs-annals-x-2-2024-113-2024","url":null,"abstract":"Abstract. Robust and automated point cloud registration methods are required in many geoscience applications using multi-temporal and multi-modal 3D point clouds. Therefore, a 3D keypoint-based coarse registration workflow has been implemented, utilizing the ISS keypoint detector and 3DSmoothNet descriptor. This paper contributes to keypoint-based registration research through variations of the standard workflow proposed in the literature, applying a two-staged strategy of global and local keypoint matching as well as prototypical keypoint projection and fine registration based on ICP. Further, by testing the utilized detector and descriptor on unstructured, multi-temporal and multi-source point clouds with variations in point cloud density, generalization ability is tested outside benchmark data. Therefore, data of the Bøverbreen glacier in Jotunheimen, Norway has been acquired in 2022 and 2023, deploying UAV-based image matching and terrestrial laser scanning. The results show good performance of the implemented robust matching algorithm PROSAC, requiring fewer iterations than the well-known RANSAC approach, but solving the rigid body transformation with TEASER++ is faster and more robust to outliers without demanding pre-knowledge of the data. Further, the results identify the keypoint detection as most limiting factor in speed and accuracy. Summarizing, keypoint-based coarse registration on low density point clouds, applying a global and local matching strategy and transformation estimation using TEASER++ is recommended. Keypoint projection shows potential, increasing number and precision in low density clouds, but has to be more robust. Further research needs to be carried out, focusing on identifying a fast and robust keypoint detector.\u0000","PeriodicalId":508124,"journal":{"name":"ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences","volume":"121 30","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141361787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-10DOI: 10.5194/isprs-annals-x-2-2024-57-2024
M. Elias, S. Isfort, A. Eltner, Hans-Gerd Maas
Abstract. The presented research investigates different strategies to acquire high-precision digital elevation models (DEMs) of complex and inaccessible terrain using Structure-from-Motion and Multi-View Stereo applied to data of an unoccupied aerial system (UAS) equipped with real-time-kinematic (RTK)-GNSS. The survey scenarios are taken from real-life situations and thus, in comparison to many previous studies, provide information on how to operate under challenging conditions in difficult terrain. Among others, the study examines the influence of different flight configurations (parallel axes and cross-grid), flight altitudes (relative to ellipsoid or terrain) and associated variations in ground sampling distance, image orientations (nadir and oblique), advanced camera self-calibration techniques and georeferencing strategies in image block processing (direct and integrated) on the overall accuracy of the resulting DEMs. Random and systematic errors, including spatial patterns such as doming and bowling, are quantified using check points and differences between DEM calculations and independently acquired surface data from laser scans. This comprehensive analysis contributes valuable insights for UAS-based analysis of complex terrain with improved accuracy in DEM generation and subsequent applications like change detection.
摘要所提交的研究调查了在复杂和难以进入的地形中使用结构-运动和多视图立体获取高精度数字高程模型（DEMs）的不同策略，这些策略应用于配备实时运动学（RTK）-全球导航卫星系统（GNSS）的无人驾驶航空系统（UAS）的数据。调查场景取自现实生活中的情况，因此，与之前的许多研究相比，它提供了如何在困难地形的挑战条件下进行操作的信息。除其他外，该研究还审查了不同飞行配置（平行轴和交叉网格）、飞行高度（相对于椭球面或地形）以及地面采样距离的相关变化、图像方向（正中线和斜线）、先进的相机自校准技术和图像块处理（直接和综合）中的地理参照策略对所生成的 DEM 整体精度的影响。随机误差和系统误差，包括穹顶和弓形等空间模式，均通过检查点和 DEM 计算结果与独立获取的激光扫描表面数据之间的差异进行量化。这项综合分析为基于无人机系统的复杂地形分析提供了宝贵的见解，提高了 DEM 生成和后续应用（如变化检测）的精度。
{"title":"UAS Photogrammetry for Precise Digital Elevation Models of Complex Topography: A Strategy Guide","authors":"M. Elias, S. Isfort, A. Eltner, Hans-Gerd Maas","doi":"10.5194/isprs-annals-x-2-2024-57-2024","DOIUrl":"https://doi.org/10.5194/isprs-annals-x-2-2024-57-2024","url":null,"abstract":"Abstract. The presented research investigates different strategies to acquire high-precision digital elevation models (DEMs) of complex and inaccessible terrain using Structure-from-Motion and Multi-View Stereo applied to data of an unoccupied aerial system (UAS) equipped with real-time-kinematic (RTK)-GNSS. The survey scenarios are taken from real-life situations and thus, in comparison to many previous studies, provide information on how to operate under challenging conditions in difficult terrain. Among others, the study examines the influence of different flight configurations (parallel axes and cross-grid), flight altitudes (relative to ellipsoid or terrain) and associated variations in ground sampling distance, image orientations (nadir and oblique), advanced camera self-calibration techniques and georeferencing strategies in image block processing (direct and integrated) on the overall accuracy of the resulting DEMs. Random and systematic errors, including spatial patterns such as doming and bowling, are quantified using check points and differences between DEM calculations and independently acquired surface data from laser scans. This comprehensive analysis contributes valuable insights for UAS-based analysis of complex terrain with improved accuracy in DEM generation and subsequent applications like change detection.\u0000","PeriodicalId":508124,"journal":{"name":"ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences","volume":"110 15","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141360869","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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：