Point cloud non local denoising using local surface descriptor similarity

Jean Emmanuel Deschaud; François Goulette

Point cloud non local denoising using local surface descriptor similarity

Jean Emmanuel Deschaud
, François Goulette

Mines ParisTech

Résultats de recherche: Contribution à un journal › Article de conférence › Revue par des pairs

Résumé

This article addresses the problem of denoising 3D data from LIDAR. It is a step often required to allow a good reconstruction of surfaces represented by point clouds. In this paper, we present an original algorithm inspired by a recent method developed by (Buades and Morel, 2005) in the field of image processing, the Non Local Denoising (NLD). With a local geometric descriptor, we look for points that have similarities in order to reduce noise while preserving the surface details. We describe local geometry by MLS surfaces and we use a local reference frame invariant by rotation for denoising points. We present our results on synthetic and real data.

langue originale	Anglais
Pages (de - à)	109-114
Nombre de pages	6
journal	International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
Volume	38
état	Publié - 1 janv. 2010
Modification externe	Oui
Evénement	ISPRS Technical Commission III Symposium on Photogrammetric Computer Vision and Image Analysis, PCV 2010 - Saint-Mande, France Durée: 1 sept. 2010 → 3 sept. 2010

Autres fichiers et liens

Lien vers la publication dans Scopus

Contient cette citation

@article{079380c017964fcc804f846f1bd00f6a,

title = "Point cloud non local denoising using local surface descriptor similarity",

abstract = "This article addresses the problem of denoising 3D data from LIDAR. It is a step often required to allow a good reconstruction of surfaces represented by point clouds. In this paper, we present an original algorithm inspired by a recent method developed by (Buades and Morel, 2005) in the field of image processing, the Non Local Denoising (NLD). With a local geometric descriptor, we look for points that have similarities in order to reduce noise while preserving the surface details. We describe local geometry by MLS surfaces and we use a local reference frame invariant by rotation for denoising points. We present our results on synthetic and real data.",

keywords = "Denoising, Descriptor, LIDAR, Mesh, Point Cloud",

author = "Deschaud, \{Jean Emmanuel\} and Fran{\c c}ois Goulette",

year = "2010",

month = jan,

day = "1",

language = "English",

volume = "38",

pages = "109--114",

journal = "International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives",

issn = "1682-1750",

note = "ISPRS Technical Commission III Symposium on Photogrammetric Computer Vision and Image Analysis, PCV 2010 ; Conference date: 01-09-2010 Through 03-09-2010",

}

Point cloud non local denoising using local surface descriptor similarity. / Deschaud, Jean Emmanuel; Goulette, François.
Dans: International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, Vol 38, 01.01.2010, p. 109-114.

Résultats de recherche: Contribution à un journal › Article de conférence › Revue par des pairs

TY - JOUR

T1 - Point cloud non local denoising using local surface descriptor similarity

AU - Deschaud, Jean Emmanuel

AU - Goulette, François

PY - 2010/1/1

Y1 - 2010/1/1

N2 - This article addresses the problem of denoising 3D data from LIDAR. It is a step often required to allow a good reconstruction of surfaces represented by point clouds. In this paper, we present an original algorithm inspired by a recent method developed by (Buades and Morel, 2005) in the field of image processing, the Non Local Denoising (NLD). With a local geometric descriptor, we look for points that have similarities in order to reduce noise while preserving the surface details. We describe local geometry by MLS surfaces and we use a local reference frame invariant by rotation for denoising points. We present our results on synthetic and real data.

AB - This article addresses the problem of denoising 3D data from LIDAR. It is a step often required to allow a good reconstruction of surfaces represented by point clouds. In this paper, we present an original algorithm inspired by a recent method developed by (Buades and Morel, 2005) in the field of image processing, the Non Local Denoising (NLD). With a local geometric descriptor, we look for points that have similarities in order to reduce noise while preserving the surface details. We describe local geometry by MLS surfaces and we use a local reference frame invariant by rotation for denoising points. We present our results on synthetic and real data.

KW - Denoising

KW - Descriptor

KW - LIDAR

KW - Mesh

KW - Point Cloud

UR - https://www.scopus.com/pages/publications/84855648541

M3 - Conference article

AN - SCOPUS:84855648541

SN - 1682-1750

VL - 38

SP - 109

EP - 114

JO - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives

JF - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives

T2 - ISPRS Technical Commission III Symposium on Photogrammetric Computer Vision and Image Analysis, PCV 2010

Y2 - 1 September 2010 through 3 September 2010

ER -

Point cloud non local denoising using local surface descriptor similarity

Résumé

Autres fichiers et liens

Empreinte digitale

Contient cette citation