Conditional random fields for urban scene classification with full waveform LiDAR data

Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review

Autorschaft

  • Joachim Niemeyer
  • Jan Dirk Wegner
  • Clément Mallet
  • Franz Rottensteiner
  • Uwe Soergel

Organisationseinheiten

Externe Organisationen

  • Université Paris-Est Créteil Val-de-Marne (UPEC)
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Titel des SammelwerksPhotogrammetric Image Analysis, ISPRS Conference, PIA 2011, Proceedings
Seiten233-244
Seitenumfang12
PublikationsstatusVeröffentlicht - 2011
VeranstaltungISPRS Conference on Photogrammetric Image Analysis, PIA 2011 - Munich, Deutschland
Dauer: 5 Okt. 2011 → 7 Okt. 2011

Publikationsreihe

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Band6952 LNCS
ISSN (Print)0302-9743
ISSN (elektronisch)1611-3349

Abstract

We propose a context-based classification method for point clouds acquired by full waveform airborne laser scanners. As these devices provide a higher point density and additional information like echo width or type of return, an accurate distinction of several object classes is possible. However, especially in dense urban areas correct labelling is a challenging task. Therefore, we incorporate context knowledge by using Conditional Random Fields. Typical object structures are learned in a training step and improve the results of the point-based classification process. We validate our approach with two real-world datasets and by a comparison to Support Vector Machines and Markov Random Fields.

ASJC Scopus Sachgebiete

Zitieren

Conditional random fields for urban scene classification with full waveform LiDAR data. / Niemeyer, Joachim; Wegner, Jan Dirk; Mallet, Clément et al.
Photogrammetric Image Analysis, ISPRS Conference, PIA 2011, Proceedings. 2011. S. 233-244 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Band 6952 LNCS).

Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review

Niemeyer, J, Wegner, JD, Mallet, C, Rottensteiner, F & Soergel, U 2011, Conditional random fields for urban scene classification with full waveform LiDAR data. in Photogrammetric Image Analysis, ISPRS Conference, PIA 2011, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Bd. 6952 LNCS, S. 233-244, ISPRS Conference on Photogrammetric Image Analysis, PIA 2011, Munich, Deutschland, 5 Okt. 2011. https://doi.org/10.1007/978-3-642-24393-6_20
Niemeyer, J., Wegner, J. D., Mallet, C., Rottensteiner, F., & Soergel, U. (2011). Conditional random fields for urban scene classification with full waveform LiDAR data. In Photogrammetric Image Analysis, ISPRS Conference, PIA 2011, Proceedings (S. 233-244). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Band 6952 LNCS). https://doi.org/10.1007/978-3-642-24393-6_20
Niemeyer J, Wegner JD, Mallet C, Rottensteiner F, Soergel U. Conditional random fields for urban scene classification with full waveform LiDAR data. in Photogrammetric Image Analysis, ISPRS Conference, PIA 2011, Proceedings. 2011. S. 233-244. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-642-24393-6_20
Niemeyer, Joachim ; Wegner, Jan Dirk ; Mallet, Clément et al. / Conditional random fields for urban scene classification with full waveform LiDAR data. Photogrammetric Image Analysis, ISPRS Conference, PIA 2011, Proceedings. 2011. S. 233-244 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Download
@inproceedings{4c07e2b83d8643d5ace6dc96baf15736,
title = "Conditional random fields for urban scene classification with full waveform LiDAR data",
abstract = "We propose a context-based classification method for point clouds acquired by full waveform airborne laser scanners. As these devices provide a higher point density and additional information like echo width or type of return, an accurate distinction of several object classes is possible. However, especially in dense urban areas correct labelling is a challenging task. Therefore, we incorporate context knowledge by using Conditional Random Fields. Typical object structures are learned in a training step and improve the results of the point-based classification process. We validate our approach with two real-world datasets and by a comparison to Support Vector Machines and Markov Random Fields.",
keywords = "3D Point Cloud, Classification, Conditional Random Fields, Full Waveform LiDAR, Urban",
author = "Joachim Niemeyer and Wegner, {Jan Dirk} and Cl{\'e}ment Mallet and Franz Rottensteiner and Uwe Soergel",
year = "2011",
doi = "10.1007/978-3-642-24393-6_20",
language = "English",
isbn = "9783642243929",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "233--244",
booktitle = "Photogrammetric Image Analysis, ISPRS Conference, PIA 2011, Proceedings",
note = "ISPRS Conference on Photogrammetric Image Analysis, PIA 2011 ; Conference date: 05-10-2011 Through 07-10-2011",

}

Download

TY - GEN

T1 - Conditional random fields for urban scene classification with full waveform LiDAR data

AU - Niemeyer, Joachim

AU - Wegner, Jan Dirk

AU - Mallet, Clément

AU - Rottensteiner, Franz

AU - Soergel, Uwe

PY - 2011

Y1 - 2011

N2 - We propose a context-based classification method for point clouds acquired by full waveform airborne laser scanners. As these devices provide a higher point density and additional information like echo width or type of return, an accurate distinction of several object classes is possible. However, especially in dense urban areas correct labelling is a challenging task. Therefore, we incorporate context knowledge by using Conditional Random Fields. Typical object structures are learned in a training step and improve the results of the point-based classification process. We validate our approach with two real-world datasets and by a comparison to Support Vector Machines and Markov Random Fields.

AB - We propose a context-based classification method for point clouds acquired by full waveform airborne laser scanners. As these devices provide a higher point density and additional information like echo width or type of return, an accurate distinction of several object classes is possible. However, especially in dense urban areas correct labelling is a challenging task. Therefore, we incorporate context knowledge by using Conditional Random Fields. Typical object structures are learned in a training step and improve the results of the point-based classification process. We validate our approach with two real-world datasets and by a comparison to Support Vector Machines and Markov Random Fields.

KW - 3D Point Cloud

KW - Classification

KW - Conditional Random Fields

KW - Full Waveform LiDAR

KW - Urban

UR - http://www.scopus.com/inward/record.url?scp=80054034092&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-24393-6_20

DO - 10.1007/978-3-642-24393-6_20

M3 - Conference contribution

AN - SCOPUS:80054034092

SN - 9783642243929

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 233

EP - 244

BT - Photogrammetric Image Analysis, ISPRS Conference, PIA 2011, Proceedings

T2 - ISPRS Conference on Photogrammetric Image Analysis, PIA 2011

Y2 - 5 October 2011 through 7 October 2011

ER -