Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


thesis_Shengxi Gui.pdf (5.23 MB)

ETD Abstract Container

Abstract Header

A Model-Driven Approach for LoD-2 Modeling Using DSM from Multi-stereo Satellite Images

Gui, Shengxi
http://orcid.org/0000-0001-8932-0577
http://rave.ohiolink.edu/etdc/view?acc_num=osu1593620776362528

Abstract Details

2020, Master of Science, Ohio State University, Civil Engineering.
Digital surface models (DSM) generated from multi-stereo satellite images are getting higher quality due to the improved data resolution and photogrammetric reconstruction algorithms. This effectively acts as a unique data source for 3D building modeling, because it provides a much wider data coverage with lower cost than the traditionally used LiDAR and airborne photogrammetry data. Although 3D building modeling from point clouds has been intensively investigated, most of the methods are still ad-hoc to specific types of buildings and require high-quality data sources as inputs. Therefore, when being applied to satellite- based point cloud or DSM, these developed approaches are not readily translatable and more adaptive and robust methods are needed. As a result, most of the existing works on building modeling from satellite DSM are to achieve LoD-1 modeling. In this paper, we propose a model-driven approach that aims to provide LoD-2 building models through a “decomposition-fitting approach”. The proposed method starts with pre-processing the generated DSM and Multispectral Orthophoto through a deep learning based object classification and a few elementary building models, and then it decomposes each 3D segments into simpler ones to be fitted by our elementary models. The decomposition and fitting take building-specific parameters such as hip lines, as well as non-rigid and regularized transformations to optimize the flexibility for using a minimal set of elementary models. Complete building models are generated in the last step by merging individually fitted elementary models. Our results on a diverse experimental dataset of city cross different continents demonstrate that our approach is able to provide LoD-2 models with medium-to-high quality.
Rongjun Qin (Advisor)
Desheng Liu (Committee Member)
Alper Yilmaz (Committee Member)
67 p.
Civil Engineering; Geographic Information Science; Remote Sensing
Building Model fitting, data-driven, decomposition and merging, multi-stereo satellite images

Recommended Citations

Citations

  • Gui, S. (2020). A Model-Driven Approach for LoD-2 Modeling Using DSM from Multi-stereo Satellite Images [Master's thesis, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1593620776362528

    APA Style (7th edition)

  • Gui, Shengxi. A Model-Driven Approach for LoD-2 Modeling Using DSM from Multi-stereo Satellite Images. 2020. Ohio State University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1593620776362528.

    MLA Style (8th edition)

  • Gui, Shengxi. "A Model-Driven Approach for LoD-2 Modeling Using DSM from Multi-stereo Satellite Images." Master's thesis, Ohio State University, 2020. http://rave.ohiolink.edu/etdc/view?acc_num=osu1593620776362528

    Chicago Manual of Style (17th edition)

osu1593620776362528
550
© 2020, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.