User-assisted Object Detection by Segment Based Similarity Measures in Mobile Laser Scanner Data

S.J. Oude Elberink, B.J. Kemboi

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

18 Citations (Scopus)

Abstract

This paper describes a method that aims to find all instances of a certain object in Mobile Laser Scanner (MLS) data. In a userassisted approach, a sample segment of an object is selected, and all similar objects are to be found. By selecting samples from multiple classes, a classification can be performed. Key assumption in this approach is that a one-to-one relationship exists between segments and objects. In this paper the focus is twofold: (1) to explain how to get proper segments, and (2) to describe how to find similar objects. Point attributes that help separating neighbouring objects are presented. These point attributes are used in an attributed connected component algorithm where segments are grown, based on proximity and attribute values. Per component, a feature vector is proposed that consists of two parts. The first is a height histogram, containing information on the height distribution of points within a component. The second contains size and shape information, based on the components’ bounding box. A simple correlation function is used to find similarities between samples, as selected by a user, and other components. Our approach is tested on a MLS dataset, containing over 300 objects in 13 classes. Detection accuracies heavily depend on the success of the segmentation, and the number of selected samples in combination with the variety of object types in the scene
Original languageEnglish
Title of host publicationISPRS Technical Commission III Symposium proceedings, 5-9 September, 2014, Zurich Switzerland. ISPRS Archives, Vol XL-3
Place of PublicationZurich
PublisherInternational Society for Photogrammetry and Remote Sensing (ISPRS)
Pages239-246
Publication statusPublished - 5 Sep 2014

Fingerprint

Lasers
Object detection

Keywords

  • METIS-305215

Cite this

Oude Elberink, S. J., & Kemboi, B. J. (2014). User-assisted Object Detection by Segment Based Similarity Measures in Mobile Laser Scanner Data. In ISPRS Technical Commission III Symposium proceedings, 5-9 September, 2014, Zurich Switzerland. ISPRS Archives, Vol XL-3 (pp. 239-246). Zurich: International Society for Photogrammetry and Remote Sensing (ISPRS).
Oude Elberink, S.J. ; Kemboi, B.J. / User-assisted Object Detection by Segment Based Similarity Measures in Mobile Laser Scanner Data. ISPRS Technical Commission III Symposium proceedings, 5-9 September, 2014, Zurich Switzerland. ISPRS Archives, Vol XL-3. Zurich : International Society for Photogrammetry and Remote Sensing (ISPRS), 2014. pp. 239-246
@inproceedings{e5bd408968a64afd836042bd8bc04baa,
title = "User-assisted Object Detection by Segment Based Similarity Measures in Mobile Laser Scanner Data",
abstract = "This paper describes a method that aims to find all instances of a certain object in Mobile Laser Scanner (MLS) data. In a userassisted approach, a sample segment of an object is selected, and all similar objects are to be found. By selecting samples from multiple classes, a classification can be performed. Key assumption in this approach is that a one-to-one relationship exists between segments and objects. In this paper the focus is twofold: (1) to explain how to get proper segments, and (2) to describe how to find similar objects. Point attributes that help separating neighbouring objects are presented. These point attributes are used in an attributed connected component algorithm where segments are grown, based on proximity and attribute values. Per component, a feature vector is proposed that consists of two parts. The first is a height histogram, containing information on the height distribution of points within a component. The second contains size and shape information, based on the components’ bounding box. A simple correlation function is used to find similarities between samples, as selected by a user, and other components. Our approach is tested on a MLS dataset, containing over 300 objects in 13 classes. Detection accuracies heavily depend on the success of the segmentation, and the number of selected samples in combination with the variety of object types in the scene",
keywords = "METIS-305215",
author = "{Oude Elberink}, S.J. and B.J. Kemboi",
year = "2014",
month = "9",
day = "5",
language = "English",
pages = "239--246",
booktitle = "ISPRS Technical Commission III Symposium proceedings, 5-9 September, 2014, Zurich Switzerland. ISPRS Archives, Vol XL-3",
publisher = "International Society for Photogrammetry and Remote Sensing (ISPRS)",

}

Oude Elberink, SJ & Kemboi, BJ 2014, User-assisted Object Detection by Segment Based Similarity Measures in Mobile Laser Scanner Data. in ISPRS Technical Commission III Symposium proceedings, 5-9 September, 2014, Zurich Switzerland. ISPRS Archives, Vol XL-3. International Society for Photogrammetry and Remote Sensing (ISPRS), Zurich, pp. 239-246.

User-assisted Object Detection by Segment Based Similarity Measures in Mobile Laser Scanner Data. / Oude Elberink, S.J.; Kemboi, B.J.

ISPRS Technical Commission III Symposium proceedings, 5-9 September, 2014, Zurich Switzerland. ISPRS Archives, Vol XL-3. Zurich : International Society for Photogrammetry and Remote Sensing (ISPRS), 2014. p. 239-246.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - User-assisted Object Detection by Segment Based Similarity Measures in Mobile Laser Scanner Data

AU - Oude Elberink, S.J.

AU - Kemboi, B.J.

PY - 2014/9/5

Y1 - 2014/9/5

N2 - This paper describes a method that aims to find all instances of a certain object in Mobile Laser Scanner (MLS) data. In a userassisted approach, a sample segment of an object is selected, and all similar objects are to be found. By selecting samples from multiple classes, a classification can be performed. Key assumption in this approach is that a one-to-one relationship exists between segments and objects. In this paper the focus is twofold: (1) to explain how to get proper segments, and (2) to describe how to find similar objects. Point attributes that help separating neighbouring objects are presented. These point attributes are used in an attributed connected component algorithm where segments are grown, based on proximity and attribute values. Per component, a feature vector is proposed that consists of two parts. The first is a height histogram, containing information on the height distribution of points within a component. The second contains size and shape information, based on the components’ bounding box. A simple correlation function is used to find similarities between samples, as selected by a user, and other components. Our approach is tested on a MLS dataset, containing over 300 objects in 13 classes. Detection accuracies heavily depend on the success of the segmentation, and the number of selected samples in combination with the variety of object types in the scene

AB - This paper describes a method that aims to find all instances of a certain object in Mobile Laser Scanner (MLS) data. In a userassisted approach, a sample segment of an object is selected, and all similar objects are to be found. By selecting samples from multiple classes, a classification can be performed. Key assumption in this approach is that a one-to-one relationship exists between segments and objects. In this paper the focus is twofold: (1) to explain how to get proper segments, and (2) to describe how to find similar objects. Point attributes that help separating neighbouring objects are presented. These point attributes are used in an attributed connected component algorithm where segments are grown, based on proximity and attribute values. Per component, a feature vector is proposed that consists of two parts. The first is a height histogram, containing information on the height distribution of points within a component. The second contains size and shape information, based on the components’ bounding box. A simple correlation function is used to find similarities between samples, as selected by a user, and other components. Our approach is tested on a MLS dataset, containing over 300 objects in 13 classes. Detection accuracies heavily depend on the success of the segmentation, and the number of selected samples in combination with the variety of object types in the scene

KW - METIS-305215

M3 - Conference contribution

SP - 239

EP - 246

BT - ISPRS Technical Commission III Symposium proceedings, 5-9 September, 2014, Zurich Switzerland. ISPRS Archives, Vol XL-3

PB - International Society for Photogrammetry and Remote Sensing (ISPRS)

CY - Zurich

ER -

Oude Elberink SJ, Kemboi BJ. User-assisted Object Detection by Segment Based Similarity Measures in Mobile Laser Scanner Data. In ISPRS Technical Commission III Symposium proceedings, 5-9 September, 2014, Zurich Switzerland. ISPRS Archives, Vol XL-3. Zurich: International Society for Photogrammetry and Remote Sensing (ISPRS). 2014. p. 239-246