Maximum likely scale estimation

Marco Loog; Kim Steenstrup Pedersen; Bo Markussen

doi:10.1007/11577812_13

Maximum likely scale estimation

Marco Loog, Kim Steenstrup Pedersen, Bo Markussen

3 Citations (Scopus)

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.

Original language	English
Title of host publication	Deep Structure, Singularities, and Computer Vision
Publisher	Springer
Publication date	2005
Pages	146-156
ISBN (Print)	978-3-540-29836-6
DOIs	https://doi.org/10.1007/11577812_13
Publication status	Published - 2005
Event	First International Workshop in Deep Structure, Singularities, and Computer Vision (DSSCV) - Maastricht, Netherlands Duration: 29 Nov 2010 → … Conference number: 1

Conference

Conference	First International Workshop in Deep Structure, Singularities, and Computer Vision (DSSCV)
Number	1
Country/Territory	Netherlands
City	Maastricht
Period	29/11/2010 → …

Series	Lecture notes in computer science
Volume	3753/2005
ISSN	0302-9743

Access to Document

10.1007/11577812_13

http://www.springerlink.de/content/yk837544xp6k8274/fulltext.pdf

Cite this

Loog, M, Pedersen, KS & Markussen, B 2005, Maximum likely scale estimation. in Deep Structure, Singularities, and Computer Vision. Springer, Lecture notes in computer science, vol. 3753/2005, pp. 146-156, First International Workshop in Deep Structure, Singularities, and Computer Vision (DSSCV), Maastricht, Netherlands, 29/11/2010. https://doi.org/10.1007/11577812_13

@inproceedings{4ca614704ce411dd8d9f000ea68e967b,

title = "Maximum likely scale estimation",

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. ",

author = "Marco Loog and Pedersen, {Kim Steenstrup} and Bo Markussen",

year = "2005",

doi = "10.1007/11577812_13",

language = "English",

isbn = "978-3-540-29836-6",

series = "Lecture notes in computer science",

publisher = "Springer",

pages = "146--156",

booktitle = "Deep Structure, Singularities, and Computer Vision",

note = "First International Workshop in Deep Structure, Singularities, and Computer Vision (DSSCV) ; Conference date: 29-11-2010",

}

TY - GEN

T1 - Maximum likely scale estimation

AU - Loog, Marco

AU - Pedersen, Kim Steenstrup

AU - Markussen, Bo

N1 - Conference code: 1

PY - 2005

Y1 - 2005

N2 - 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.

AB - 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.

U2 - 10.1007/11577812_13

DO - 10.1007/11577812_13

M3 - Article in proceedings

SN - 978-3-540-29836-6

T3 - Lecture notes in computer science

SP - 146

EP - 156

BT - Deep Structure, Singularities, and Computer Vision

PB - Springer

T2 - First International Workshop in Deep Structure, Singularities, and Computer Vision (DSSCV)

Y2 - 29 November 2010

ER -