Regular texture analysis as statistical model selection

Junwei Han; Stephen J. McKenna; Ruixuan Wang

doi:10.1007/978-3-540-88693-8_18

Regular texture analysis as statistical model selection

Junwei Han, Stephen J. McKenna, Ruixuan Wang

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

13 Citations (Scopus)

Abstract

An approach to the analysis of images of regular texture is proposed in which lattice hypotheses are used to define statistical models. These models are then compared in terms of their ability to explain the image. A method based on this approach is described in which lattice hypotheses are generated using analysis of peaks in the image autocorrelation function, statistical models are based on Gaussian or Gaussian mixture clusters, and model comparison is performed using the marginal likelihood as approximated by the Bayes Information Criterion (BIC). Experiments on public domain regular texture images and a commercial textile image archive demonstrate substantially improved accuracy compared to two competing methods. The method is also used for classification of texture images as regular or irregular. An application to thumbnail image extraction is discussed.

Original language	English
Title of host publication	Computer Vision – ECCV 2008
Subtitle of host publication	10th European Conference on Computer Vision Marseille, France, October 12-18, 2008. Proceedings, Part IV
Editors	David Forsyth, Philip Torr, Andrew Zisserman
Place of Publication	Berlin
Publisher	Springer
Pages	242-255
Number of pages	14
ISBN (Electronic)	9783540886938
ISBN (Print)	9783540886921
DOIs	https://doi.org/10.1007/978-3-540-88693-8_18
Publication status	Published - 2008
Event	10th European Conference on Computer Vision - Palais des Congrès Parc Chanot, Marseille, France Duration: 12 Oct 2008 → 18 Oct 2008 http://eccv2008.inrialpes.fr/

Publication series

Name	Lecture notes in computer science
Publisher	Springer
Volume	5305
ISSN (Print)	0302-9743

Conference

Conference	10th European Conference on Computer Vision
Abbreviated title	ECCV 2008
Country/Territory	France
City	Marseille
Period	12/10/08 → 18/10/08
Internet address	http://eccv2008.inrialpes.fr/

Keywords

Periodicity

Access to Document

10.1007/978-3-540-88693-8_18

Cite this

Han, J., McKenna, S. J., & Wang, R. (2008). Regular texture analysis as statistical model selection. In D. Forsyth, P. Torr, & A. Zisserman (Eds.), Computer Vision – ECCV 2008: 10th European Conference on Computer Vision Marseille, France, October 12-18, 2008. Proceedings, Part IV (pp. 242-255). (Lecture notes in computer science; Vol. 5305). Springer . https://doi.org/10.1007/978-3-540-88693-8_18

Han, Junwei ; McKenna, Stephen J. ; Wang, Ruixuan. / Regular texture analysis as statistical model selection. Computer Vision – ECCV 2008: 10th European Conference on Computer Vision Marseille, France, October 12-18, 2008. Proceedings, Part IV. editor / David Forsyth ; Philip Torr ; Andrew Zisserman. Berlin : Springer , 2008. pp. 242-255 (Lecture notes in computer science).

@inproceedings{d3b7eac5923d40bc80a3f42b7acb512f,

title = "Regular texture analysis as statistical model selection",

abstract = "An approach to the analysis of images of regular texture is proposed in which lattice hypotheses are used to define statistical models. These models are then compared in terms of their ability to explain the image. A method based on this approach is described in which lattice hypotheses are generated using analysis of peaks in the image autocorrelation function, statistical models are based on Gaussian or Gaussian mixture clusters, and model comparison is performed using the marginal likelihood as approximated by the Bayes Information Criterion (BIC). Experiments on public domain regular texture images and a commercial textile image archive demonstrate substantially improved accuracy compared to two competing methods. The method is also used for classification of texture images as regular or irregular. An application to thumbnail image extraction is discussed.",

keywords = "Periodicity",

author = "Junwei Han and McKenna, {Stephen J.} and Ruixuan Wang",

year = "2008",

doi = "10.1007/978-3-540-88693-8_18",

language = "English",

isbn = "9783540886921",

series = "Lecture notes in computer science",

publisher = "Springer ",

pages = "242--255",

editor = "David Forsyth and Philip Torr and Andrew Zisserman",

booktitle = "Computer Vision – ECCV 2008",

note = "10th European Conference on Computer Vision , ECCV 2008 ; Conference date: 12-10-2008 Through 18-10-2008",

url = "http://eccv2008.inrialpes.fr/",

}

Han, J, McKenna, SJ & Wang, R 2008, Regular texture analysis as statistical model selection. in D Forsyth, P Torr & A Zisserman (eds), Computer Vision – ECCV 2008: 10th European Conference on Computer Vision Marseille, France, October 12-18, 2008. Proceedings, Part IV. Lecture notes in computer science, vol. 5305, Springer , Berlin, pp. 242-255, 10th European Conference on Computer Vision , Marseille, France, 12/10/08. https://doi.org/10.1007/978-3-540-88693-8_18

Regular texture analysis as statistical model selection. / Han, Junwei; McKenna, Stephen J.; Wang, Ruixuan.
Computer Vision – ECCV 2008: 10th European Conference on Computer Vision Marseille, France, October 12-18, 2008. Proceedings, Part IV. ed. / David Forsyth; Philip Torr; Andrew Zisserman. Berlin: Springer , 2008. p. 242-255 (Lecture notes in computer science; Vol. 5305).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Regular texture analysis as statistical model selection

AU - Han, Junwei

AU - McKenna, Stephen J.

AU - Wang, Ruixuan

PY - 2008

Y1 - 2008

N2 - An approach to the analysis of images of regular texture is proposed in which lattice hypotheses are used to define statistical models. These models are then compared in terms of their ability to explain the image. A method based on this approach is described in which lattice hypotheses are generated using analysis of peaks in the image autocorrelation function, statistical models are based on Gaussian or Gaussian mixture clusters, and model comparison is performed using the marginal likelihood as approximated by the Bayes Information Criterion (BIC). Experiments on public domain regular texture images and a commercial textile image archive demonstrate substantially improved accuracy compared to two competing methods. The method is also used for classification of texture images as regular or irregular. An application to thumbnail image extraction is discussed.

AB - An approach to the analysis of images of regular texture is proposed in which lattice hypotheses are used to define statistical models. These models are then compared in terms of their ability to explain the image. A method based on this approach is described in which lattice hypotheses are generated using analysis of peaks in the image autocorrelation function, statistical models are based on Gaussian or Gaussian mixture clusters, and model comparison is performed using the marginal likelihood as approximated by the Bayes Information Criterion (BIC). Experiments on public domain regular texture images and a commercial textile image archive demonstrate substantially improved accuracy compared to two competing methods. The method is also used for classification of texture images as regular or irregular. An application to thumbnail image extraction is discussed.

KW - Periodicity

U2 - 10.1007/978-3-540-88693-8_18

DO - 10.1007/978-3-540-88693-8_18

M3 - Conference contribution

SN - 9783540886921

T3 - Lecture notes in computer science

SP - 242

EP - 255

BT - Computer Vision – ECCV 2008

A2 - Forsyth, David

A2 - Torr, Philip

A2 - Zisserman, Andrew

PB - Springer

CY - Berlin

T2 - 10th European Conference on Computer Vision

Y2 - 12 October 2008 through 18 October 2008

ER -

Han J, McKenna SJ, Wang R. Regular texture analysis as statistical model selection. In Forsyth D, Torr P, Zisserman A, editors, Computer Vision – ECCV 2008: 10th European Conference on Computer Vision Marseille, France, October 12-18, 2008. Proceedings, Part IV. Berlin: Springer . 2008. p. 242-255. (Lecture notes in computer science). doi: 10.1007/978-3-540-88693-8_18

Regular texture analysis as statistical model selection

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this