Determining Gabor-filter parameters for texture segmentation

Dennis F. Dunn, William Evan Higgins, Joseph Wakeley

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Citations (Scopus)

Abstract

The ability to segment a textured image into separate regions (texture segmentation) continues to be a challenging problem in computer vision. Many texture-segmentation schemes are based on a filter-bank model, where the filters (henceforth referred to as Gabor Filters) are derived from Gabor elementary functions. The goal of these methods is to transform texture differences into detectable filter-output discontinuities at texture boundaries. Then, one can segment the image into differently textured regions. Distinct discontinuities occur, however, only if the parameters defining the Gabor filters are suitably chosen. Some previous analysis has shown how to design appropriate filters for discriminating simple textures. Designing filters for more general textures, though, has largely been done ad hoc. We have devised a new, more effective, more rigorously based method for determining Gabor-filter parameters. The method is based on an exhaustive, but efficient, search of Gabor-filter parameter space and on a detection-theory formulation of a Gabor filter's output. We provide qualitative arguments and experimental results indicating that our new method is more effective than other methods in producing suitable filter parameters. We demonstrate that our model also gives good filter designs for a variety of texture types.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
PublisherPubl by Int Soc for Optical Engineering
Pages51-63
Number of pages13
Volume1826
ISBN (Print)0819410276
StatePublished - Jan 1 1993
EventIntelligent Robots and Computer Vision XI: Biological, Neural Net, and 3-D Methods - Boston, MA, USA
Duration: Nov 18 1992Nov 20 1992

Other

OtherIntelligent Robots and Computer Vision XI: Biological, Neural Net, and 3-D Methods
CityBoston, MA, USA
Period11/18/9211/20/92

Fingerprint

Gabor filters
Texture Segmentation
Gabor Filter
Texture
textures
Textures
filters
Filter
Filter Design
Discontinuity
Region Segmentation
discontinuity
Elementary Functions
Filter Banks
Output
Computer Vision
Parameter Space
output
Filter banks
Continue

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Cite this

Dunn, D. F., Higgins, W. E., & Wakeley, J. (1993). Determining Gabor-filter parameters for texture segmentation. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 1826, pp. 51-63). Publ by Int Soc for Optical Engineering.
Dunn, Dennis F. ; Higgins, William Evan ; Wakeley, Joseph. / Determining Gabor-filter parameters for texture segmentation. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 1826 Publ by Int Soc for Optical Engineering, 1993. pp. 51-63
@inproceedings{1bf551f06a664eec97aeed3ff93795ee,
title = "Determining Gabor-filter parameters for texture segmentation",
abstract = "The ability to segment a textured image into separate regions (texture segmentation) continues to be a challenging problem in computer vision. Many texture-segmentation schemes are based on a filter-bank model, where the filters (henceforth referred to as Gabor Filters) are derived from Gabor elementary functions. The goal of these methods is to transform texture differences into detectable filter-output discontinuities at texture boundaries. Then, one can segment the image into differently textured regions. Distinct discontinuities occur, however, only if the parameters defining the Gabor filters are suitably chosen. Some previous analysis has shown how to design appropriate filters for discriminating simple textures. Designing filters for more general textures, though, has largely been done ad hoc. We have devised a new, more effective, more rigorously based method for determining Gabor-filter parameters. The method is based on an exhaustive, but efficient, search of Gabor-filter parameter space and on a detection-theory formulation of a Gabor filter's output. We provide qualitative arguments and experimental results indicating that our new method is more effective than other methods in producing suitable filter parameters. We demonstrate that our model also gives good filter designs for a variety of texture types.",
author = "Dunn, {Dennis F.} and Higgins, {William Evan} and Joseph Wakeley",
year = "1993",
month = "1",
day = "1",
language = "English (US)",
isbn = "0819410276",
volume = "1826",
pages = "51--63",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",
publisher = "Publ by Int Soc for Optical Engineering",

}

Dunn, DF, Higgins, WE & Wakeley, J 1993, Determining Gabor-filter parameters for texture segmentation. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 1826, Publ by Int Soc for Optical Engineering, pp. 51-63, Intelligent Robots and Computer Vision XI: Biological, Neural Net, and 3-D Methods, Boston, MA, USA, 11/18/92.

Determining Gabor-filter parameters for texture segmentation. / Dunn, Dennis F.; Higgins, William Evan; Wakeley, Joseph.

Proceedings of SPIE - The International Society for Optical Engineering. Vol. 1826 Publ by Int Soc for Optical Engineering, 1993. p. 51-63.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Determining Gabor-filter parameters for texture segmentation

AU - Dunn, Dennis F.

AU - Higgins, William Evan

AU - Wakeley, Joseph

PY - 1993/1/1

Y1 - 1993/1/1

N2 - The ability to segment a textured image into separate regions (texture segmentation) continues to be a challenging problem in computer vision. Many texture-segmentation schemes are based on a filter-bank model, where the filters (henceforth referred to as Gabor Filters) are derived from Gabor elementary functions. The goal of these methods is to transform texture differences into detectable filter-output discontinuities at texture boundaries. Then, one can segment the image into differently textured regions. Distinct discontinuities occur, however, only if the parameters defining the Gabor filters are suitably chosen. Some previous analysis has shown how to design appropriate filters for discriminating simple textures. Designing filters for more general textures, though, has largely been done ad hoc. We have devised a new, more effective, more rigorously based method for determining Gabor-filter parameters. The method is based on an exhaustive, but efficient, search of Gabor-filter parameter space and on a detection-theory formulation of a Gabor filter's output. We provide qualitative arguments and experimental results indicating that our new method is more effective than other methods in producing suitable filter parameters. We demonstrate that our model also gives good filter designs for a variety of texture types.

AB - The ability to segment a textured image into separate regions (texture segmentation) continues to be a challenging problem in computer vision. Many texture-segmentation schemes are based on a filter-bank model, where the filters (henceforth referred to as Gabor Filters) are derived from Gabor elementary functions. The goal of these methods is to transform texture differences into detectable filter-output discontinuities at texture boundaries. Then, one can segment the image into differently textured regions. Distinct discontinuities occur, however, only if the parameters defining the Gabor filters are suitably chosen. Some previous analysis has shown how to design appropriate filters for discriminating simple textures. Designing filters for more general textures, though, has largely been done ad hoc. We have devised a new, more effective, more rigorously based method for determining Gabor-filter parameters. The method is based on an exhaustive, but efficient, search of Gabor-filter parameter space and on a detection-theory formulation of a Gabor filter's output. We provide qualitative arguments and experimental results indicating that our new method is more effective than other methods in producing suitable filter parameters. We demonstrate that our model also gives good filter designs for a variety of texture types.

UR - http://www.scopus.com/inward/record.url?scp=0027148493&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027148493&partnerID=8YFLogxK

M3 - Conference contribution

SN - 0819410276

VL - 1826

SP - 51

EP - 63

BT - Proceedings of SPIE - The International Society for Optical Engineering

PB - Publ by Int Soc for Optical Engineering

ER -

Dunn DF, Higgins WE, Wakeley J. Determining Gabor-filter parameters for texture segmentation. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 1826. Publ by Int Soc for Optical Engineering. 1993. p. 51-63