Detection of object wings in fused range and intensity imagery

Greg Chungmou Lee, George Stockman

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

We present methods for detecting primitives in scenes containing objects whose surfaces are approximately piecewise quadric using a sensor that yields fused range and intensity. The primitives detected in a fused image are viewpoint dependent and are called object wings, which consist of adjacent fragments of object contour and surface. Wings are small enough so that some object parts can be sensed despite significant occlusion, yet rich enough to be of use in indexing to candidate models, in constraining object pose, and in aggregation methods which combine primitives into larger structures or objects. Fitting procedures are defined for specific wing primitives. Experiments with both synthetic and real data show that classification within the wing types of planar, spherical, conical, cylindrical can be achieved with over 90% accuracy. Contour information available in fitting fused data improves fitting relative to use of surface data alone. Moreover, it is shown that only a small number of contour and surface samples are needed for extraction of primitives.

Original languageEnglish
Pages (from-to)137-158
Number of pages22
JournalPattern Recognition
Volume31
Issue number2
Publication statusPublished - 1998 Feb 1

Fingerprint

Agglomeration
Sensors
Experiments

Keywords

  • 3D recognition
  • 3D representation
  • Fused image
  • Model fitting
  • Recognition-by-parts
  • Shape
  • Shape classification
  • Shape primitives

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

Detection of object wings in fused range and intensity imagery. / Lee, Greg Chungmou; Stockman, George.

In: Pattern Recognition, Vol. 31, No. 2, 01.02.1998, p. 137-158.

Research output: Contribution to journalArticle

Lee, Greg Chungmou ; Stockman, George. / Detection of object wings in fused range and intensity imagery. In: Pattern Recognition. 1998 ; Vol. 31, No. 2. pp. 137-158.
@article{aba1dceceadc4c9e86c4d7a72739819c,
title = "Detection of object wings in fused range and intensity imagery",
abstract = "We present methods for detecting primitives in scenes containing objects whose surfaces are approximately piecewise quadric using a sensor that yields fused range and intensity. The primitives detected in a fused image are viewpoint dependent and are called object wings, which consist of adjacent fragments of object contour and surface. Wings are small enough so that some object parts can be sensed despite significant occlusion, yet rich enough to be of use in indexing to candidate models, in constraining object pose, and in aggregation methods which combine primitives into larger structures or objects. Fitting procedures are defined for specific wing primitives. Experiments with both synthetic and real data show that classification within the wing types of planar, spherical, conical, cylindrical can be achieved with over 90{\%} accuracy. Contour information available in fitting fused data improves fitting relative to use of surface data alone. Moreover, it is shown that only a small number of contour and surface samples are needed for extraction of primitives.",
keywords = "3D recognition, 3D representation, Fused image, Model fitting, Recognition-by-parts, Shape, Shape classification, Shape primitives",
author = "Lee, {Greg Chungmou} and George Stockman",
year = "1998",
month = "2",
day = "1",
language = "English",
volume = "31",
pages = "137--158",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "2",

}

TY - JOUR

T1 - Detection of object wings in fused range and intensity imagery

AU - Lee, Greg Chungmou

AU - Stockman, George

PY - 1998/2/1

Y1 - 1998/2/1

N2 - We present methods for detecting primitives in scenes containing objects whose surfaces are approximately piecewise quadric using a sensor that yields fused range and intensity. The primitives detected in a fused image are viewpoint dependent and are called object wings, which consist of adjacent fragments of object contour and surface. Wings are small enough so that some object parts can be sensed despite significant occlusion, yet rich enough to be of use in indexing to candidate models, in constraining object pose, and in aggregation methods which combine primitives into larger structures or objects. Fitting procedures are defined for specific wing primitives. Experiments with both synthetic and real data show that classification within the wing types of planar, spherical, conical, cylindrical can be achieved with over 90% accuracy. Contour information available in fitting fused data improves fitting relative to use of surface data alone. Moreover, it is shown that only a small number of contour and surface samples are needed for extraction of primitives.

AB - We present methods for detecting primitives in scenes containing objects whose surfaces are approximately piecewise quadric using a sensor that yields fused range and intensity. The primitives detected in a fused image are viewpoint dependent and are called object wings, which consist of adjacent fragments of object contour and surface. Wings are small enough so that some object parts can be sensed despite significant occlusion, yet rich enough to be of use in indexing to candidate models, in constraining object pose, and in aggregation methods which combine primitives into larger structures or objects. Fitting procedures are defined for specific wing primitives. Experiments with both synthetic and real data show that classification within the wing types of planar, spherical, conical, cylindrical can be achieved with over 90% accuracy. Contour information available in fitting fused data improves fitting relative to use of surface data alone. Moreover, it is shown that only a small number of contour and surface samples are needed for extraction of primitives.

KW - 3D recognition

KW - 3D representation

KW - Fused image

KW - Model fitting

KW - Recognition-by-parts

KW - Shape

KW - Shape classification

KW - Shape primitives

UR - http://www.scopus.com/inward/record.url?scp=0031995539&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031995539&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0031995539

VL - 31

SP - 137

EP - 158

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 2

ER -