Advanced

Constructing a neural system for surface inspection

Grunditz, Carl-Henrik; Walder, Martin and Spaanenburg, Lambert LU (2004) Joint SAIS/SSLS Workshop, 2004 In SAIS Workshop p.68-73
Abstract
Visual quality assurance techniques focus on the detection and qualification of abnormal structures in the image of an object. The features of abnormality are extracted through image mining, whereupon classification is performed on characteristic combinations. Many techniques for feature extraction have been proposed, but the feed-forward neural network is seldom utilized despite its popularity in other application areas. Based on this wide experience base, this paper shows how a multi-tier feed-forward network can be constructed to model detectable peaks using only the physical properties of the image domain. This generic architecture can easily be adapted for different applications, as in metal plate inspection and protein detection,... (More)
Visual quality assurance techniques focus on the detection and qualification of abnormal structures in the image of an object. The features of abnormality are extracted through image mining, whereupon classification is performed on characteristic combinations. Many techniques for feature extraction have been proposed, but the feed-forward neural network is seldom utilized despite its popularity in other application areas. Based on this wide experience base, this paper shows how a multi-tier feed-forward network can be constructed to model detectable peaks using only the physical properties of the image domain. This generic architecture can easily be adapted for different applications, as in metal plate inspection and protein detection, with mean error rate below 5%. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
in
SAIS Workshop
editor
Malec, Jacek
pages
68 - 73
publisher
SAIS
conference name
Joint SAIS/SSLS Workshop, 2004
external identifiers
  • Scopus:10844252252
language
English
LU publication?
yes
id
1297a587-f844-4da1-b21f-035953ebc573 (old id 1028969)
date added to LUP
2008-02-11 10:58:40
date last changed
2017-01-01 08:04:45
@inproceedings{1297a587-f844-4da1-b21f-035953ebc573,
  abstract     = {Visual quality assurance techniques focus on the detection and qualification of abnormal structures in the image of an object. The features of abnormality are extracted through image mining, whereupon classification is performed on characteristic combinations. Many techniques for feature extraction have been proposed, but the feed-forward neural network is seldom utilized despite its popularity in other application areas. Based on this wide experience base, this paper shows how a multi-tier feed-forward network can be constructed to model detectable peaks using only the physical properties of the image domain. This generic architecture can easily be adapted for different applications, as in metal plate inspection and protein detection, with mean error rate below 5%.},
  author       = {Grunditz, Carl-Henrik and Walder, Martin and Spaanenburg, Lambert},
  booktitle    = {SAIS Workshop},
  editor       = {Malec, Jacek},
  language     = {eng},
  pages        = {68--73},
  publisher    = {SAIS},
  title        = {Constructing a neural system for surface inspection},
  year         = {2004},
}