| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 62.39 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
The need for a relevant viable approach to fit point clouds obtained by 3D laser scanning, to a desirable surface, has been object ofa substantial research effort and progresso in the past two decades in a wide range of scientific and technological fields. However, this task is far from being a trivial task. First, because of the randomness of the sampled points obtained, which in most cases count with additional noise points. Secondly, in point clouds it is frequent to find lacks ofdata, leading to the existence ofholes in the surface. As far as it is possible to know, all the methods used to achieve the fitting surfaces, present diferente undesirable behaviours, under different conditions. In the present work we present a hybrid method to reconstruct the surfaces associated to synthetic point clouds randomly generated. Parametric studies are carried out to illustrate and characterize the performance ofthe different techniques implemented..
Description
Keywords
Points cloud Laser scanning Surface reconstruction Meshless method Compactly-supported radial functions Partition ofunity
Citation
BERNARDO, G. M. S.; LOJA, M. A. R. – Reconstruction of surfaces from unstructured points clouds, using compactly-supported radial basis functions. In SYMCOMP 2015. Faro, Portugal: ECCOMAS, 2015. ISBN 978-989-96264-6-1. Pp. 427-454.