Abstract
We introduce an algorithm which emphasizes edges that form continuous curves. The algorithm avoids binary decisions, simply complementing edge information with confidence values. Our algorithm is based on Mallat's method of edge detection. Continuity is detected by the combination of responses of Mallat filters of different scales. Due to its structure the algorithm can profitably be used as part of a modular computer vision system.
Supported by a grant from the German Federal Ministry for Science and Technology.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Jan C. Vorbrüggen. Zwei Modelle zur datengetriebenen Segmentierung visueller Daten, volume 47 of Reihe Physik. Verlag Harri Deutsch, Thun, Frankfurt am Main, 1995.
C. Eckes. Integration daten-und modellgetriebener Segmentierung in natürlichen Bildern. Diplomarbeit, Universität Dortmund, 1995.
Christian Eckes and Jan C. Vorbrüggen. Combining data-driven and model-based cues for segmentation of video sequences. World Congress on Neural Networks, San Diego, USA, 1996.
D. Marr. Early processing of visual information. Phil. Trans. R. Soc. Lond. B, 275:483–524, 1976.
D. Marr and E. Hildreth. Theory of edge detection. Proc. R. Soc. Lond. B, 207:187–217, 1980.
D. Marr, T.Poggio, and S.Ullman. Bandpass channels, zero-crossings, and early visual information processing. J. Opt. Soc. Am., 69:914–916, 1979.
S. Mallat. A theory for multiresolution signal decomposition: The wavelet representation. IEEE Trans. PAMI, 11:674–693, 1987.
S.Mallat and S.Zhong. Characterization of signals from multiscale edges. Technical report, Courant Institute, NYU, New York, USA, 1991.
Rakesh Mohan and Ramakant Nevatia. Using perceptual organization to extract 3-d structures. IEEE Trans. PAMI, 11(11):1121–1138, 1989.
P. Parent and S. W. Zucker. Trace inference, curvature consistency and curve detection. Technical Report CIM-86-3, Computer Vision and Robotic Lab., McGill University, 1986.
S. W. Zucker, R. Hummel, and A. Rosenfeld. An application of relaxation labeling to line and curve enhancement. IEEE Trans. on Computers, 26:394–403, 1977.
D. Marr. Vision. W.H. Freeman and Co, San Francisco, first edition, 1982.
E. Kefalea and C.v.d. Malsburg. Object classification based on contours with Elastic Graph Matching. (work in progress).
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Rehse, O., Pötzsch, M., von der Malsburg, C. (1996). Edge information: A confidence based algorithm emphazising continuous curves. In: von der Malsburg, C., von Seelen, W., Vorbrüggen, J.C., Sendhoff, B. (eds) Artificial Neural Networks — ICANN 96. ICANN 1996. Lecture Notes in Computer Science, vol 1112. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-61510-5_143
Download citation
DOI: https://doi.org/10.1007/3-540-61510-5_143
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-61510-1
Online ISBN: 978-3-540-68684-2
eBook Packages: Springer Book Archive