Constructing a neural system for surface inspection

• 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%

Nyckelord

image mining

feature extraction

image classification

object image detection

visual quality assurance techniques

surface inspection

neural system

protein detection

multiple tier feedforward neural network

metal plate inspection

TEKNIKVETENSKAP

TECHNOLOGY

Publikations- och innehållstyp

kon (ämneskategori)

ref (ämneskategori)