A bound on mutual information for image registration

Mark B. Skouson, Quji Quo, Zhi-Pei Liang

Research output: Contribution to journalLetter

Abstract

An upper bound is derived for the mutual information between a fixed image and a deformable template containing a fixed number of gray-levels. The bound can be calculated by maximizing the entropy of the template under the constraint that the conditional entropy of the template, given the fixed image, be zero. This bound provides useful insight into the properties of mutual information as a similarity metric for deformable image registration. Specifically, it indicates that maximizing mutual information may not necessarily produce an optimal solution when the deformable transform is too flexible.

Original languageEnglish (US)
Pages (from-to)843-845
Number of pages3
JournalIEEE Transactions on Medical Imaging
Volume20
Issue number8
DOIs
StatePublished - Aug 1 2001

Fingerprint

Image registration
Entropy

Keywords

  • Deformation
  • Information bound
  • Mutual information
  • Registration

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology
  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Computational Theory and Mathematics

Cite this

A bound on mutual information for image registration. / Skouson, Mark B.; Quo, Quji; Liang, Zhi-Pei.

In: IEEE Transactions on Medical Imaging, Vol. 20, No. 8, 01.08.2001, p. 843-845.

Research output: Contribution to journalLetter

Skouson, Mark B. ; Quo, Quji ; Liang, Zhi-Pei. / A bound on mutual information for image registration. In: IEEE Transactions on Medical Imaging. 2001 ; Vol. 20, No. 8. pp. 843-845.
@article{d821bfa650c845b3abb95d3d1de6bf4f,
title = "A bound on mutual information for image registration",
abstract = "An upper bound is derived for the mutual information between a fixed image and a deformable template containing a fixed number of gray-levels. The bound can be calculated by maximizing the entropy of the template under the constraint that the conditional entropy of the template, given the fixed image, be zero. This bound provides useful insight into the properties of mutual information as a similarity metric for deformable image registration. Specifically, it indicates that maximizing mutual information may not necessarily produce an optimal solution when the deformable transform is too flexible.",
keywords = "Deformation, Information bound, Mutual information, Registration",
author = "Skouson, {Mark B.} and Quji Quo and Zhi-Pei Liang",
year = "2001",
month = "8",
day = "1",
doi = "10.1109/42.938252",
language = "English (US)",
volume = "20",
pages = "843--845",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "8",

}

TY - JOUR

T1 - A bound on mutual information for image registration

AU - Skouson, Mark B.

AU - Quo, Quji

AU - Liang, Zhi-Pei

PY - 2001/8/1

Y1 - 2001/8/1

N2 - An upper bound is derived for the mutual information between a fixed image and a deformable template containing a fixed number of gray-levels. The bound can be calculated by maximizing the entropy of the template under the constraint that the conditional entropy of the template, given the fixed image, be zero. This bound provides useful insight into the properties of mutual information as a similarity metric for deformable image registration. Specifically, it indicates that maximizing mutual information may not necessarily produce an optimal solution when the deformable transform is too flexible.

AB - An upper bound is derived for the mutual information between a fixed image and a deformable template containing a fixed number of gray-levels. The bound can be calculated by maximizing the entropy of the template under the constraint that the conditional entropy of the template, given the fixed image, be zero. This bound provides useful insight into the properties of mutual information as a similarity metric for deformable image registration. Specifically, it indicates that maximizing mutual information may not necessarily produce an optimal solution when the deformable transform is too flexible.

KW - Deformation

KW - Information bound

KW - Mutual information

KW - Registration

UR - http://www.scopus.com/inward/record.url?scp=0035413319&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035413319&partnerID=8YFLogxK

U2 - 10.1109/42.938252

DO - 10.1109/42.938252

M3 - Letter

C2 - 11513035

AN - SCOPUS:0035413319

VL - 20

SP - 843

EP - 845

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 8

ER -