Abstract
In this paper we evaluate the performance with respect to classification of red blood cells of an invariant statistical classifier that was successfully applied to a variety of object recognition tasks. The classifier is based on distance functions invariant to affine transformations and additive brightness and on kernel densities within a Bayesian framework. Given a database of 5062 grayscale images, we follow an’ appearance based’ approach obtaining an error rate of 16.3%, lying below the human error rate of greater than 20%. Our experiments show the general applicability of the approach taken. A comparison with results obtained in other domains underlines the task dependency of the performance of different classification algorithms.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
B. Deuticke, R. Grebe, and C. Haest. Action of Drugs on the Erythrocyte Membrane. In J. Harris, editor, Blood Cell Biochemistry, volume 1. Plenum Press, New York, pages 475–529, 1990.
D. Keysers, J. Dahmen, T. Theiner, and H. Ney. Experiments with an Extended Tangent Distance. In Proceedings 15th International Conference on Pattern Recognition, volume 2, Barcelona, Spain, pages 38–42, September 2000.
J. Dahmen, J. Hektor, R. Perrey, and H. Ney. Automatic Classification of Red Blood Cells using Gaussian Mixture Densities. In Bildverarbeitung für die Medizin, München, pages 331–335, March 2000.
T. Fischer. Department of Physiology, RWTH Aachen University of Technology, personal communication, 1999.
M. Schönfeld and R. Grebe. Automatic Shape Quantification of Freely Suspended Red Blood Cells by Isodensity Contour Tracing and Tangent Counting. Computer Methods and Programs in Biomedicine, 28:217–224, 1989.
J. Dahmen, D. Keysers, M. Motter, H. Ney, T. Lehmann, and B. Wein. An Automatic Approach to Invariant Radiograph Classification. In Bildverarbeitung für die Medizin, Springer, Lubeck, Germany, March 2001. This volume.
P. Simard, Y. Le Cun, J. Denker, and B. Victorri. Transformation Invariance in Pattern Recognition — Tangent Distance and Tangent Propagation. In G. Orr and K.-R. Muller, editors, Neural networks: tricks of the trade, volume 1524 of Lecture Notes in Computer Science, Springer, Heidelberg, pages 239–274, 1998.
D. Keysers, J. Dahmen, and H. Ney. A Probabilistic View on Tangent Distance. In 22. DAGM Symposium Mustererkennung 2000, Springer, Kiel, Germany, pages 107–114, September 2000.
J. Dahmen, D. Keysers, M. O. Güld, and H. Ney. Invariant Image Object Recognition using Mixture Densities. In Proceedings 15th International Conference on Pattern Recognition, volume 2, Barcelona, Spain, pages 614–617, September 2000.
B. Schölkopf, P. Simard, A. Smola, and V. Vapnik. Prior Knowledge in Support Vector Kernels. In M. I. Jordan, M. J. Kearns, and S. A. Solla, editors, Advances in Neural Inf. Proc. Systems, volume 10. MIT Press, pages 640–646, 1998.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Keysers, D., Dahmen, J., Ney, H. (2001). Invariant Classification of Red Blood Cells. In: Handels, H., Horsch, A., Lehmann, T., Meinzer, HP. (eds) Bildverarbeitung für die Medizin 2001. Informatik aktuell. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-56714-8_68
Download citation
DOI: https://doi.org/10.1007/978-3-642-56714-8_68
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-41690-6
Online ISBN: 978-3-642-56714-8
eBook Packages: Springer Book Archive