Extracting printed designs and woven patterns from textile images

Wei Jia; Stephen J. McKenna; Annette A. Ward

Extracting printed designs and woven patterns from textile images

Wei Jia
, Stephen J. McKenna
, Annette A. Ward

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

Abstract

The extraction of printed designs and woven patterns from textiles is formulated as a pixel labelling problem. Algorithms based on Markov random field (MRF) optimisation and reestimation are described and evaluated on images from an historical fabric archive. A method for quantitative evaluation is presented and used to compare the performance of MRF models optimised using alpha-expansion and iterated conditional modes, both with and without parameter reestimation. Results are promising for potential application to content-based indexing and browsing.

Original language	English
Title of host publication	VISAPP 2009
Subtitle of host publication	proceedings of the Fourth International Conference on Computer Vision Theory and Applications, Lisboa, Portugal, February 5-8, 2009
Editors	Alpesh Ranchordas, Helder Araujo
Place of Publication	Setubal
Publisher	Institute for Systems and Technologies of Information, Control and Communication
Pages	201-208
Number of pages	8
Volume	1
ISBN (Print)	9789898111692
Publication status	Published - 2009
Event	4th International Conference on Computer Vision Theory and Applications - Lisboa, Lisbon, Portugal Duration: 5 Feb 2009 → 8 Feb 2009 http://www.visapp.visigrapp.org/VISAPP2009/

Conference

Conference	4th International Conference on Computer Vision Theory and Applications
Abbreviated title	VISAPP 2009
Country/Territory	Portugal
City	Lisbon
Period	5/02/09 → 8/02/09
Internet address	http://www.visapp.visigrapp.org/VISAPP2009/

Keywords

Textile segmentation
Pixel labelling
Markov random field
Quantitative evaluation
Energy minimization
Segmentation

Cite this

Jia, W., McKenna, S. J., & Ward, A. A. (2009). Extracting printed designs and woven patterns from textile images. In A. Ranchordas, & H. Araujo (Eds.), VISAPP 2009: proceedings of the Fourth International Conference on Computer Vision Theory and Applications, Lisboa, Portugal, February 5-8, 2009 (Vol. 1, pp. 201-208). Institute for Systems and Technologies of Information, Control and Communication.

Jia, Wei ; McKenna, Stephen J. ; Ward, Annette A. / Extracting printed designs and woven patterns from textile images. VISAPP 2009: proceedings of the Fourth International Conference on Computer Vision Theory and Applications, Lisboa, Portugal, February 5-8, 2009 . editor / Alpesh Ranchordas ; Helder Araujo. Vol. 1 Setubal : Institute for Systems and Technologies of Information, Control and Communication, 2009. pp. 201-208

@inproceedings{2d94787ef37f4adb9775da923486645f,

title = "Extracting printed designs and woven patterns from textile images",

abstract = "The extraction of printed designs and woven patterns from textiles is formulated as a pixel labelling problem. Algorithms based on Markov random field (MRF) optimisation and reestimation are described and evaluated on images from an historical fabric archive. A method for quantitative evaluation is presented and used to compare the performance of MRF models optimised using alpha-expansion and iterated conditional modes, both with and without parameter reestimation. Results are promising for potential application to content-based indexing and browsing.",

keywords = "Textile segmentation, Pixel labelling, Markov random field, Quantitative evaluation, Energy minimization, Segmentation",

author = "Wei Jia and McKenna, \{Stephen J.\} and Ward, \{Annette A.\}",

year = "2009",

language = "English",

isbn = "9789898111692",

volume = "1",

pages = "201--208",

editor = "Alpesh Ranchordas and Helder Araujo",

booktitle = "VISAPP 2009",

publisher = "Institute for Systems and Technologies of Information, Control and Communication",

note = "4th International Conference on Computer Vision Theory and Applications, VISAPP 2009 ; Conference date: 05-02-2009 Through 08-02-2009",

url = "http://www.visapp.visigrapp.org/VISAPP2009/",

}

Jia, W, McKenna, SJ & Ward, AA 2009, Extracting printed designs and woven patterns from textile images. in A Ranchordas & H Araujo (eds), VISAPP 2009: proceedings of the Fourth International Conference on Computer Vision Theory and Applications, Lisboa, Portugal, February 5-8, 2009 . vol. 1, Institute for Systems and Technologies of Information, Control and Communication, Setubal, pp. 201-208, 4th International Conference on Computer Vision Theory and Applications, Lisbon, Portugal, 5/02/09.

Extracting printed designs and woven patterns from textile images. / Jia, Wei; McKenna, Stephen J.; Ward, Annette A.
VISAPP 2009: proceedings of the Fourth International Conference on Computer Vision Theory and Applications, Lisboa, Portugal, February 5-8, 2009 . ed. / Alpesh Ranchordas; Helder Araujo. Vol. 1 Setubal: Institute for Systems and Technologies of Information, Control and Communication, 2009. p. 201-208.

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

TY - GEN

T1 - Extracting printed designs and woven patterns from textile images

AU - Jia, Wei

AU - McKenna, Stephen J.

AU - Ward, Annette A.

PY - 2009

Y1 - 2009

N2 - The extraction of printed designs and woven patterns from textiles is formulated as a pixel labelling problem. Algorithms based on Markov random field (MRF) optimisation and reestimation are described and evaluated on images from an historical fabric archive. A method for quantitative evaluation is presented and used to compare the performance of MRF models optimised using alpha-expansion and iterated conditional modes, both with and without parameter reestimation. Results are promising for potential application to content-based indexing and browsing.

AB - The extraction of printed designs and woven patterns from textiles is formulated as a pixel labelling problem. Algorithms based on Markov random field (MRF) optimisation and reestimation are described and evaluated on images from an historical fabric archive. A method for quantitative evaluation is presented and used to compare the performance of MRF models optimised using alpha-expansion and iterated conditional modes, both with and without parameter reestimation. Results are promising for potential application to content-based indexing and browsing.

KW - Textile segmentation

KW - Pixel labelling

KW - Markov random field

KW - Quantitative evaluation

KW - Energy minimization

KW - Segmentation

M3 - Conference contribution

SN - 9789898111692

VL - 1

SP - 201

EP - 208

BT - VISAPP 2009

A2 - Ranchordas, Alpesh

A2 - Araujo, Helder

PB - Institute for Systems and Technologies of Information, Control and Communication

CY - Setubal

T2 - 4th International Conference on Computer Vision Theory and Applications

Y2 - 5 February 2009 through 8 February 2009

ER -

Jia W, McKenna SJ, Ward AA. Extracting printed designs and woven patterns from textile images. In Ranchordas A, Araujo H, editors, VISAPP 2009: proceedings of the Fourth International Conference on Computer Vision Theory and Applications, Lisboa, Portugal, February 5-8, 2009 . Vol. 1. Setubal: Institute for Systems and Technologies of Information, Control and Communication. 2009. p. 201-208

Extracting printed designs and woven patterns from textile images

Abstract

Conference

Keywords

Fingerprint

Cite this