Share Email Print

Proceedings Paper

Optimal Affine-Invariant Point Matching
Author(s): Mauro Costa; Robert Haralick; Tsaiyun Phillips; Linda Shapiro
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

The affine-transformation matching scheme proposed by Hummel and Wolfson (1988) is very efficient in a model-based matching system, not only in terms of the computational complexity involved, but also in terms of the simplicity of the method. This paper addresses the implementation of the affine-invariant point matching, applied to the problem of recognizing and determining the pose of sheet metal parts. It points out errors that can occur with this method due to quantization, stability, symmetry, and noise problems. By beginning with an explicit noise model which the Hummel and Wolfson technique lacks, we can derive an optimal approach which overcomes these problems. We show that results obtained with the new algorithm are clearly better than the results from the original method.

Paper Details

Date Published: 21 March 1989
PDF: 17 pages
Proc. SPIE 1095, Applications of Artificial Intelligence VII, (21 March 1989); doi: 10.1117/12.969300
Show Author Affiliations
Mauro Costa, University of Washington (United States)
Robert Haralick, University of Washington (United States)
Tsaiyun Phillips, University of Washington (United States)
Linda Shapiro, University of Washington (United States)

Published in SPIE Proceedings Vol. 1095:
Applications of Artificial Intelligence VII
Mohan M. Trivedi, Editor(s)

© SPIE. Terms of Use
Back to Top