Choosing the number of labels in image segmentation

Yonghuai Liu; J. H. Draper; Alan Philip Gay; Catherine Howarth; R. R. Martin

doi:10.1109/ETFA.2011.6059090

Choosing the number of labels in image segmentation

Yonghuai Liu, J. H. Draper, Alan Philip Gay, Catherine Howarth, R. R. Martin

Research output: Contribution to conference › Paper

Abstract

Image segmentation is a fundamental yet challenging step during image analysis. In this paper we propose a novel method for choosing the number of labels during automatic image segmentation. It minimizes an objective function based on the number of labels, the segmentation errors, and consistency of labels between neighboring pixels. An experimental study on representative data shows encouraging results.

Original language	English
DOIs	https://doi.org/10.1109/ETFA.2011.6059090
Publication status	Published - 2011
Event	Emerging Technologies and Factory Automation - Toulouse, France Duration: 05 Sept 2011 → 09 Sept 2011 Conference number: 16

Conference

Conference	Emerging Technologies and Factory Automation
Abbreviated title	ETFA-2011
Country/Territory	France
City	Toulouse
Period	05 Sept 2011 → 09 Sept 2011

Access to Document

10.1109/ETFA.2011.6059090

Permanent link

Persistent link

Cite this

@conference{d74d47884d664fd1b1e315b4f77335db,

title = "Choosing the number of labels in image segmentation",

abstract = "Image segmentation is a fundamental yet challenging step during image analysis. In this paper we propose a novel method for choosing the number of labels during automatic image segmentation. It minimizes an objective function based on the number of labels, the segmentation errors, and consistency of labels between neighboring pixels. An experimental study on representative data shows encouraging results.",

author = "Yonghuai Liu and Draper, {J. H.} and Gay, {Alan Philip} and Catherine Howarth and Martin, {R. R.}",

year = "2011",

doi = "10.1109/ETFA.2011.6059090",

language = "English",

note = "Emerging Technologies and Factory Automation, ETFA-2011 ; Conference date: 05-09-2011 Through 09-09-2011",

}

TY - CONF

T1 - Choosing the number of labels in image segmentation

AU - Liu, Yonghuai

AU - Draper, J. H.

AU - Gay, Alan Philip

AU - Howarth, Catherine

AU - Martin, R. R.

N1 - Conference code: 16

PY - 2011

Y1 - 2011

N2 - Image segmentation is a fundamental yet challenging step during image analysis. In this paper we propose a novel method for choosing the number of labels during automatic image segmentation. It minimizes an objective function based on the number of labels, the segmentation errors, and consistency of labels between neighboring pixels. An experimental study on representative data shows encouraging results.

AB - Image segmentation is a fundamental yet challenging step during image analysis. In this paper we propose a novel method for choosing the number of labels during automatic image segmentation. It minimizes an objective function based on the number of labels, the segmentation errors, and consistency of labels between neighboring pixels. An experimental study on representative data shows encouraging results.

UR - http://hdl.handle.net/2160/10390

U2 - 10.1109/ETFA.2011.6059090

DO - 10.1109/ETFA.2011.6059090

M3 - Paper

T2 - Emerging Technologies and Factory Automation

Y2 - 5 September 2011 through 9 September 2011

ER -

Choosing the number of labels in image segmentation

Abstract

Conference

Access to Document

Permanent link

Other files and links

Fingerprint

Cite this