Share Email Print
cover

Proceedings Paper

Unsupervised image segmentation with the self-organizing map and statistical methods
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

In this paper a special type of image segmentation, a two- class segmentation, is considered. Defect detection in quality control applications is a typical two-class problem. The main idea in this paper is to train the two-class classifier with fault-free samples that is an unexpected approach. The reason is that defects are rare and expensive. The proposed defect detection is based on the following idea: an unknown sample is classified as a defect if it differs enough from the estimated prototypes of fault-free samples. The self-organizing map is used to estimate these prototypes. Surface images are used to demonstrate the proposed image segmentation procedure.

Paper Details

Date Published: 6 October 1998
PDF: 11 pages
Proc. SPIE 3522, Intelligent Robots and Computer Vision XVII: Algorithms, Techniques, and Active Vision, (6 October 1998); doi: 10.1117/12.325796
Show Author Affiliations
Jukka Iivarinen, Helsinki Univ. of Technology (Finland)
Ari J. E. Visa, Lappeenranta Univ. of Technology (Finland)


Published in SPIE Proceedings Vol. 3522:
Intelligent Robots and Computer Vision XVII: Algorithms, Techniques, and Active Vision
David P. Casasent, Editor(s)

© SPIE. Terms of Use
Back to Top