Title:
A semi-automated method of building element retrieval from point cloud
A semi-automated method of building element retrieval from point cloud
| dc.contributor.advisor | Cho, Yong Kwon | |
| dc.contributor.author | Zeng, Shiqin | |
| dc.contributor.committeeMember | Tsai, Yi-Chang James | |
| dc.contributor.committeeMember | Marks, Eric | |
| dc.contributor.department | Civil and Environmental Engineering | |
| dc.date.accessioned | 2021-06-10T16:48:42Z | |
| dc.date.available | 2021-06-10T16:48:42Z | |
| dc.date.created | 2020-05 | |
| dc.date.issued | 2020-04-28 | |
| dc.date.submitted | May 2020 | |
| dc.date.updated | 2021-06-10T16:48:42Z | |
| dc.description.abstract | 3D point cloud data can be utilized for site inspection and reverse engineering of building models. However, conventional methods for building element retrieval require a database of 3D CAD or BIM models which are unsuitable for the case of historical buildings without as-planned models or temporary structures that are not in the pre-built model. Thus, this paper proposes a semi-automated method to efficiently retrieve duplicate building elements without these constraints. First, the point cloud is processed with a pre-trained deep feature extractor to generate a 50-dimensional feature vector for each point. Next, the point cloud is segmented through feature clustering and region growing algorithms, then displayed on a user interface for selection. Lastly, the selected exemplar is provided as input to a peak-finding algorithm to determine positive matches. The results show the proposed method gets the average rates above 90% of precision and recall scores of each point cloud dataset. The proposed method can distinguish the correct building elements form the similarly-shaped candidates and complex building elements. In terms of the applicability, the study shows the proposed method has a certain tolerance of error with different selected instances or boundaries of the selected exemplar and voxel grid resolution. On the other hand, the actual computation time is reasonably fast and efficient. | |
| dc.description.degree | M.S. | |
| dc.format.mimetype | application/pdf | |
| dc.identifier.uri | http://hdl.handle.net/1853/64652 | |
| dc.language.iso | en_US | |
| dc.publisher | Georgia Institute of Technology | |
| dc.subject | Laser scanning | |
| dc.subject | point cloud | |
| dc.subject | building elements retrieval | |
| dc.subject | temporary structures | |
| dc.subject | historical buildings | |
| dc.subject | machine learning | |
| dc.subject | deep neural network | |
| dc.subject | transfer learning | |
| dc.title | A semi-automated method of building element retrieval from point cloud | |
| dc.type | Text | |
| dc.type.genre | Thesis | |
| dspace.entity.type | Publication | |
| local.contributor.advisor | Cho, Yong Kwon | |
| local.contributor.corporatename | School of Civil and Environmental Engineering | |
| local.contributor.corporatename | College of Engineering | |
| relation.isAdvisorOfPublication | edb06af8-870c-40db-8fc2-f5b417b89165 | |
| relation.isOrgUnitOfPublication | 88639fad-d3ae-4867-9e7a-7c9e6d2ecc7c | |
| relation.isOrgUnitOfPublication | 7c022d60-21d5-497c-b552-95e489a06569 | |
| thesis.degree.level | Masters |