Linear discriminants and image quality

Harrison H Barrett, T. Gooley, K. Girodias, J. Rolland, T. White, J. Yao

Research output: Contribution to journalArticle

44 Citations (Scopus)

Abstract

The use of linear discriminant functions, and particularly a discriminant function derived from the work of Harold Hotelling1, as a means of assessing image quality is reviewed. The relevant theory of ideal or Bayesian observers is briefly reviewed, and the circumstances under which this observer reduces to a linear discriminant are discussed. The Hotelling observer is suggested as a linear discriminant in more general circumstances, where the ideal observer is nonlinear and usually very difficult to calculate. Methods of calculation of the Hotelling discriminant and the associated figure of merit, the Hotelling trace, are discussed. Psychophysical studies carried out at the University of Arizona to test the predictive value of the Hotelling observer are reviewed, and it is concluded that the Hotelling model is quite useful as a predictive tool unless there are high-pass noise correlations introduced by post-processing of the images. In that case, we suggest that the Hotelling observer be modified to include spatial-frequencyselective channels analogous to those in the visual system.

Original languageEnglish (US)
Pages (from-to)451-460
Number of pages10
JournalImage and Vision Computing
Volume10
Issue number6
DOIs
StatePublished - 1992

Fingerprint

Image quality
Processing

Keywords

  • Hotelling trace
  • ideal observer
  • image quality
  • linear discriminant functions
  • medical imaging

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Barrett, H. H., Gooley, T., Girodias, K., Rolland, J., White, T., & Yao, J. (1992). Linear discriminants and image quality. Image and Vision Computing, 10(6), 451-460. https://doi.org/10.1016/0262-8856(92)90030-7

Linear discriminants and image quality. / Barrett, Harrison H; Gooley, T.; Girodias, K.; Rolland, J.; White, T.; Yao, J.

In: Image and Vision Computing, Vol. 10, No. 6, 1992, p. 451-460.

Research output: Contribution to journalArticle

Barrett, HH, Gooley, T, Girodias, K, Rolland, J, White, T & Yao, J 1992, 'Linear discriminants and image quality', Image and Vision Computing, vol. 10, no. 6, pp. 451-460. https://doi.org/10.1016/0262-8856(92)90030-7
Barrett, Harrison H ; Gooley, T. ; Girodias, K. ; Rolland, J. ; White, T. ; Yao, J. / Linear discriminants and image quality. In: Image and Vision Computing. 1992 ; Vol. 10, No. 6. pp. 451-460.
@article{260fda5f088d44b28704bc26ac1a7411,
title = "Linear discriminants and image quality",
abstract = "The use of linear discriminant functions, and particularly a discriminant function derived from the work of Harold Hotelling1, as a means of assessing image quality is reviewed. The relevant theory of ideal or Bayesian observers is briefly reviewed, and the circumstances under which this observer reduces to a linear discriminant are discussed. The Hotelling observer is suggested as a linear discriminant in more general circumstances, where the ideal observer is nonlinear and usually very difficult to calculate. Methods of calculation of the Hotelling discriminant and the associated figure of merit, the Hotelling trace, are discussed. Psychophysical studies carried out at the University of Arizona to test the predictive value of the Hotelling observer are reviewed, and it is concluded that the Hotelling model is quite useful as a predictive tool unless there are high-pass noise correlations introduced by post-processing of the images. In that case, we suggest that the Hotelling observer be modified to include spatial-frequencyselective channels analogous to those in the visual system.",
keywords = "Hotelling trace, ideal observer, image quality, linear discriminant functions, medical imaging",
author = "Barrett, {Harrison H} and T. Gooley and K. Girodias and J. Rolland and T. White and J. Yao",
year = "1992",
doi = "10.1016/0262-8856(92)90030-7",
language = "English (US)",
volume = "10",
pages = "451--460",
journal = "Image and Vision Computing",
issn = "0262-8856",
publisher = "Elsevier Limited",
number = "6",

}

TY - JOUR

T1 - Linear discriminants and image quality

AU - Barrett, Harrison H

AU - Gooley, T.

AU - Girodias, K.

AU - Rolland, J.

AU - White, T.

AU - Yao, J.

PY - 1992

Y1 - 1992

N2 - The use of linear discriminant functions, and particularly a discriminant function derived from the work of Harold Hotelling1, as a means of assessing image quality is reviewed. The relevant theory of ideal or Bayesian observers is briefly reviewed, and the circumstances under which this observer reduces to a linear discriminant are discussed. The Hotelling observer is suggested as a linear discriminant in more general circumstances, where the ideal observer is nonlinear and usually very difficult to calculate. Methods of calculation of the Hotelling discriminant and the associated figure of merit, the Hotelling trace, are discussed. Psychophysical studies carried out at the University of Arizona to test the predictive value of the Hotelling observer are reviewed, and it is concluded that the Hotelling model is quite useful as a predictive tool unless there are high-pass noise correlations introduced by post-processing of the images. In that case, we suggest that the Hotelling observer be modified to include spatial-frequencyselective channels analogous to those in the visual system.

AB - The use of linear discriminant functions, and particularly a discriminant function derived from the work of Harold Hotelling1, as a means of assessing image quality is reviewed. The relevant theory of ideal or Bayesian observers is briefly reviewed, and the circumstances under which this observer reduces to a linear discriminant are discussed. The Hotelling observer is suggested as a linear discriminant in more general circumstances, where the ideal observer is nonlinear and usually very difficult to calculate. Methods of calculation of the Hotelling discriminant and the associated figure of merit, the Hotelling trace, are discussed. Psychophysical studies carried out at the University of Arizona to test the predictive value of the Hotelling observer are reviewed, and it is concluded that the Hotelling model is quite useful as a predictive tool unless there are high-pass noise correlations introduced by post-processing of the images. In that case, we suggest that the Hotelling observer be modified to include spatial-frequencyselective channels analogous to those in the visual system.

KW - Hotelling trace

KW - ideal observer

KW - image quality

KW - linear discriminant functions

KW - medical imaging

UR - http://www.scopus.com/inward/record.url?scp=38249012152&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=38249012152&partnerID=8YFLogxK

U2 - 10.1016/0262-8856(92)90030-7

DO - 10.1016/0262-8856(92)90030-7

M3 - Article

AN - SCOPUS:38249012152

VL - 10

SP - 451

EP - 460

JO - Image and Vision Computing

JF - Image and Vision Computing

SN - 0262-8856

IS - 6

ER -