Abstract
A maximum likelihood local scale estimation principle is presented. An actual implementation of the estimation principle uses second order moments of multiple measurements at a fixed location in the image. These measurements consist of Gaussian derivatives possibly taken at several scales and/or having different derivative orders.
Although the principle is applicable to a wide variety of image models, the main focus here is on the Brownian model and its use for scale selection in natural images. Furthermore, in the examples provided, the simplifying assumption is made that the behavior of the measurements is completely characterized by all moments up to second order.
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
Pedersen, K.S.: Properties of brownian image models in scale-space. In: Griffin, L.D., Lillholm, M. (eds.) Scale-Space 2003. LNCS, vol. 2695, pp. 281–296. Springer, Heidelberg (2003)
Markussen, B., Pedersen, K.S., Loog, M.: A scale invariant covariance structure on jet space. In: Fogh Olsen, O., Florack, L.M.J., Kuijper, A. (eds.) DSSCV 2005. LNCS, vol. 3753, pp. 12–23. Springer, Heidelberg (2005)
Lindeberg, T.: Feature detection with automatic scale selection. International Journal of Computer Vision 30, 79–116 (1998)
Pedersen, K.S., Nielsen, M.: The Hausdorff dimension and scale-space normalisation of natural images. Journal of Visual Communication and Image Representation 11, 266–277 (2000)
Majer, P.: A Statistical Approach to Feature Detection and Scale Selection in Images. PhD thesis, University of Göttingen (2000)
Bouman, C.A., Sauer, K.: Maximum likelihood scale estimation for a class of Markov random fields. In: Proceedings of ICASSP 1994. IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 537–540. IEEE, Los Alamitos (1994)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Loog, M., Pedersen, K.S., Markussen, B. (2005). Maximum Likely Scale Estimation. In: Fogh Olsen, O., Florack, L., Kuijper, A. (eds) Deep Structure, Singularities, and Computer Vision. DSSCV 2005. Lecture Notes in Computer Science, vol 3753. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11577812_13
Download citation
DOI: https://doi.org/10.1007/11577812_13
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-29836-6
Online ISBN: 978-3-540-32097-5
eBook Packages: Computer ScienceComputer Science (R0)