Learning-based method to recognize and localize glassware using laser range images

Stefan Toemoe; Nageswara S.V. Rao; Reinhold C. Mann

doi:10.1016/0262-8856(95)01046-7

Learning-based method to recognize and localize glassware using laser range images

Stefan Toemoe, Nageswara S.V. Rao, Reinhold C. Mann

Autonomous and Complex Systems

Research output: Contribution to journal › Article › peer-review

2 Scopus citations

Abstract

A system that can be trained to recognize and localize glassware located on a workbench using laser range images is described. The training data consists of laser range images of objects of known classification. The image is preprocessed to isolate a box of pixels corresponding to the object, and then the box is given as an input to a neural network. The range readings from the laser range system deviate significantly from the actual distances to the glassware, and consequently a surface fit to the readings has very little resemblance to the actual glassware surface. Thus, the straightfoward method of fitting a surface to the images and matching it with the surface of known glassware is not feasible. A first version of the system has been developed based on a system of perceptrons, and the tests have been successful on the images taken by a PERCEPTRON P5000 laser range finder.

Original language	English
Pages (from-to)	131-134
Number of pages	4
Journal	Image and Vision Computing
Volume	14
Issue number	2
DOIs	https://doi.org/10.1016/0262-8856(95)01046-7
State	Published - Mar 1996

Funding

Research sponsored by the Engineering Research Program of the Office of Basic Energy Sciences, of the U.S. Department of Energy, under Contract No. DE-AC0584OR21400 with Martin Marietta Energy Systems, Inc. ’ Consultant to Oak Ridge National Laboratory. This invention was made with US Government support under subcontract No. 80X-SN93lV awarded by Martin Marietta Energy Systems, Inc., acting under contract DE-AC05-840R21400 with the US Department of Energy.

Keywords

Glassware
Laser range images
Learning
Neural networks
Perceptrons

Access to Document

10.1016/0262-8856(95)01046-7

Cite this

@article{c3cfa0b2b8b648c181a207850bffc6e2,

title = "Learning-based method to recognize and localize glassware using laser range images",

abstract = "A system that can be trained to recognize and localize glassware located on a workbench using laser range images is described. The training data consists of laser range images of objects of known classification. The image is preprocessed to isolate a box of pixels corresponding to the object, and then the box is given as an input to a neural network. The range readings from the laser range system deviate significantly from the actual distances to the glassware, and consequently a surface fit to the readings has very little resemblance to the actual glassware surface. Thus, the straightfoward method of fitting a surface to the images and matching it with the surface of known glassware is not feasible. A first version of the system has been developed based on a system of perceptrons, and the tests have been successful on the images taken by a PERCEPTRON P5000 laser range finder.",

keywords = "Glassware, Laser range images, Learning, Neural networks, Perceptrons",

author = "Stefan Toemoe and Rao, {Nageswara S.V.} and Mann, {Reinhold C.}",

year = "1996",

month = mar,

doi = "10.1016/0262-8856(95)01046-7",

language = "English",

volume = "14",

pages = "131--134",

journal = "Image and Vision Computing",

issn = "0262-8856",

publisher = "Elsevier",

number = "2",

}

TY - JOUR

T1 - Learning-based method to recognize and localize glassware using laser range images

AU - Toemoe, Stefan

AU - Rao, Nageswara S.V.

AU - Mann, Reinhold C.

PY - 1996/3

Y1 - 1996/3

N2 - A system that can be trained to recognize and localize glassware located on a workbench using laser range images is described. The training data consists of laser range images of objects of known classification. The image is preprocessed to isolate a box of pixels corresponding to the object, and then the box is given as an input to a neural network. The range readings from the laser range system deviate significantly from the actual distances to the glassware, and consequently a surface fit to the readings has very little resemblance to the actual glassware surface. Thus, the straightfoward method of fitting a surface to the images and matching it with the surface of known glassware is not feasible. A first version of the system has been developed based on a system of perceptrons, and the tests have been successful on the images taken by a PERCEPTRON P5000 laser range finder.

AB - A system that can be trained to recognize and localize glassware located on a workbench using laser range images is described. The training data consists of laser range images of objects of known classification. The image is preprocessed to isolate a box of pixels corresponding to the object, and then the box is given as an input to a neural network. The range readings from the laser range system deviate significantly from the actual distances to the glassware, and consequently a surface fit to the readings has very little resemblance to the actual glassware surface. Thus, the straightfoward method of fitting a surface to the images and matching it with the surface of known glassware is not feasible. A first version of the system has been developed based on a system of perceptrons, and the tests have been successful on the images taken by a PERCEPTRON P5000 laser range finder.

KW - Glassware

KW - Laser range images

KW - Learning

KW - Neural networks

KW - Perceptrons

UR - http://www.scopus.com/inward/record.url?scp=0030103696&partnerID=8YFLogxK

U2 - 10.1016/0262-8856(95)01046-7

DO - 10.1016/0262-8856(95)01046-7

M3 - Article

AN - SCOPUS:0030103696

SN - 0262-8856

VL - 14

SP - 131

EP - 134

JO - Image and Vision Computing

JF - Image and Vision Computing

IS - 2

ER -

Learning-based method to recognize and localize glassware using laser range images

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this