Keysers, D.
Dahmen, J.
Theiner, T.
Ney, H.
Lehrstuhl fur Inf. VI, Tech. Hochschule Aachen , Germany;
This paper appears in: Pattern Recognition, 2000. Proceedings. 15th International Conference on Publication Date: 3-7 Sept 2000
Volume: 2
On page(s):
38
- 42 vol.2
Meeting Date: 09/03/2000
- 09/07/2000
Location: Barcelona
ISBN: 0-7695-0750-6
Digital Object Identifier: 10.1109/ICPR.2000.906014
Current Version Published: 2002-08-06
Abstract
Invariance is an important aspect in image object recognition. We present results obtained with an extended tangent distance incorporated in a kernel density based Bayesian classifier to compensate for affine image variations. An image distortion model for local variations is introduced and its relationship to tangent distance is considered. The proposed classification algorithms are evaluated on databases of different domains. An excellent result of 2.2% error rate on the original USPS handwritten digits recognition task is obtained. On a database of radiographs from daily routine, best results are obtained by combining the tangent distance and the proposed distortion model
Index
Terms
Available to subscribers and IEEE members.
References
Available to subscribers and IEEE members.
Citing Documents
Available to subscribers and IEEE members.
You are not
logged in.
Guests
may access Abstract records free of charge.
Cart
