Summary: Workshop on Texture Analysis in Machine Vision, June 14-15, 1999, Oulu, Finland 1
Automatic Detection of Errors on Textures Using
Invariant Grey Scale Features and
Polynomial Classiers
Marc Schael and Hans Burkhardt
Institute for Pattern Recognition and Image Processing
Computer Science Department, University of Freiburg
Am Flughafen 17, D-79110 Freiburg, Germany
{schael,burkhardt}@informatik.uni-freiburg.de
Abstract. In this paper we propose two methods for the automatic detection of errors
on non-stochastic textures. Both methods are based on invariant grey scale features and
may be distinguished by their global or local approach, respectively. Classication of the
non-linear invariant features is done by a polynomial classier of third degree. Our test
application for the evaluation of the invariant features is the error detection on textile
surfaces. Experimental results based on the image database TILDA are presented and
discussed in this contribution.
1 Introduction
Over the last decade the automatic detection of errors on textures has gained increas-
ing importance for industrial application. Manufacturing of high quality products needs
control. In many areas of production the quality can be assessed by analyzing planar

Source: Albert-Ludwigs-Universität Freiburg, Institut für Informatik,, Lehrstuhls für Mustererkennung und Bildverarbeitung

Collections: Computer Technologies and Information Sciences