Maximum likely scale estimation

Authors:
Marco Loog;Kim Steenstrup Pedersen;Bo Markussen
Affiliations:
Image Analysis Group, Department of Innovation, IT University of Copenhagen, Denmark;Image Analysis Group, Department of Innovation, IT University of Copenhagen, Denmark;Department of Computer Science, University of Copenhagen, Denmark
Venue:
DSSCV'05 Proceedings of the First international conference on Deep Structure, Singularities, and Computer Vision
Year:
2005

Citing 3
Cited 3

Feature Detection with Automatic Scale Selection

International Journal of Computer Vision
Properties of Brownian image models in scale-space

Scale Space'03 Proceedings of the 4th international conference on Scale space methods in computer vision
A scale invariant covariance structure on jet space

DSSCV'05 Proceedings of the First international conference on Deep Structure, Singularities, and Computer Vision

Second Order Structure of Scale-Space Measurements

Journal of Mathematical Imaging and Vision
Generic maximum likely scale selection

SSVM'07 Proceedings of the 1st international conference on Scale space and variational methods in computer vision
Scale selection for supervised image segmentation

Image and Vision Computing

Quantified Score

Hi-index	0.00

Visualization

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.