Szczegóły publikacji
Opis bibliograficzny
Comparison of the filtering methods in cleaning data obtained from laser scanner / Anna PIĘTA, Krzysztof Klocek // W: SGEM 2015 : informatics, geoinformatics and remote sensing : 15th international multidisciplinary scientific geoconference : 18–24, June, 2015. Albena, Bulgaria : conference proceedings. Vol. 1, Informatics, geoinformatics, photogrammetry and remote sensing. — Sofia : STEF92 Technology Ltd., cop. 2015. — (International Multidisciplinary Scientific GeoConference SGEM ; ISSN 1314-2704). — ISBN: 978-619-7105-34-6. — S. 541–545. — Bibliogr. s. 545, Abstr.
Autorzy (2)
- AGHFranczyk Anna
- Klocek Krzysztof
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 90260 |
|---|---|
| Data dodania do BaDAP | 2015-07-21 |
| Rok publikacji | 2015 |
| Typ publikacji | materiały konferencyjne (aut.) |
| Otwarty dostęp |  |
| Konferencja | 15th international multidisciplinary scientific geoconference |
| Czasopismo/seria | International Multidisciplinary Scientific GeoConference SGEM |
Abstract
3D laser scanning is a modern, innovative technique that allows to obtain 3D model of scanned object by using controlled laser beams. Using laser beam together with precise angle measurements allows to acquire data called "point clouds". Point cloud represents shapes of scanned objects saved as points with their coordinates and intensity of the beam's reflections for a given point. The main issue, which troubles using these datasets to further analysis is presence of various types of noise which has to be remove during processing stage. There are several methods that can be used to clean point cloud. The most effective and commonly used filtering method are based on statistical relationships calculated for all analyzed points. The main disadvantages of using statistical based filtering procedures is their long computational time caused by the huge number of measurements of a given point cloud. Parallel computing and utilization of the graphical processing unit which typically handles computation only for computer graphics can significantly shorten the computation time of filtering procedures. In this paper we present the comparison of the implementation of the various filtering methods to the given point clouds that uses parallel computing. The article focuses not only on the analysis of the differences occurring in the data set after implementation of an appropriate method of filtration but also on the computational efficiency gained by utilization of the graphical processing units.
