Efficient Gabor filter design for texture segmentation

Thomas P. Weldon, William Evan Higgins, Dennis F. Dunn

Research output: Contribution to journalArticle

208 Citations (Scopus)

Abstract

Gabor filters have been successfully applied to a broad range of image processing tasks. The present paper considers the design of a single filter to segment a two-texture image. A new efficient algorithm for Gabor-filter design is presented, along with methods for estimating filter output statistics. The algorithm draws upon previous results that showed that the output of a Gabor-filtered texture is modeled well by a Rician distribution. A measure of the total output power is used to select the center frequency of the filter and is used to estimate the Rician statistics of the Gabor-filtered image. The method is further generalized to include the statistics of postfiltered outputs that are generated by a Gaussian filtering operation following the Gabor filter. The new method typically requires an order of magnitude less computation to design a filter than a previously proposed method. Experimental results demonstrate the efficacy of the method.

Original languageEnglish (US)
Pages (from-to)2005-2015
Number of pages11
JournalPattern Recognition
Volume29
Issue number12
DOIs
StatePublished - Dec 1 1996

Fingerprint

Gabor filters
Textures
Statistics
Image processing

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

Weldon, Thomas P. ; Higgins, William Evan ; Dunn, Dennis F. / Efficient Gabor filter design for texture segmentation. In: Pattern Recognition. 1996 ; Vol. 29, No. 12. pp. 2005-2015.
@article{ac4e418ee3a74a6b8622cc592bcec843,
title = "Efficient Gabor filter design for texture segmentation",
abstract = "Gabor filters have been successfully applied to a broad range of image processing tasks. The present paper considers the design of a single filter to segment a two-texture image. A new efficient algorithm for Gabor-filter design is presented, along with methods for estimating filter output statistics. The algorithm draws upon previous results that showed that the output of a Gabor-filtered texture is modeled well by a Rician distribution. A measure of the total output power is used to select the center frequency of the filter and is used to estimate the Rician statistics of the Gabor-filtered image. The method is further generalized to include the statistics of postfiltered outputs that are generated by a Gaussian filtering operation following the Gabor filter. The new method typically requires an order of magnitude less computation to design a filter than a previously proposed method. Experimental results demonstrate the efficacy of the method.",
author = "Weldon, {Thomas P.} and Higgins, {William Evan} and Dunn, {Dennis F.}",
year = "1996",
month = "12",
day = "1",
doi = "10.1016/S0031-3203(96)00047-7",
language = "English (US)",
volume = "29",
pages = "2005--2015",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "12",

}

Efficient Gabor filter design for texture segmentation. / Weldon, Thomas P.; Higgins, William Evan; Dunn, Dennis F.

In: Pattern Recognition, Vol. 29, No. 12, 01.12.1996, p. 2005-2015.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Efficient Gabor filter design for texture segmentation

AU - Weldon, Thomas P.

AU - Higgins, William Evan

AU - Dunn, Dennis F.

PY - 1996/12/1

Y1 - 1996/12/1

N2 - Gabor filters have been successfully applied to a broad range of image processing tasks. The present paper considers the design of a single filter to segment a two-texture image. A new efficient algorithm for Gabor-filter design is presented, along with methods for estimating filter output statistics. The algorithm draws upon previous results that showed that the output of a Gabor-filtered texture is modeled well by a Rician distribution. A measure of the total output power is used to select the center frequency of the filter and is used to estimate the Rician statistics of the Gabor-filtered image. The method is further generalized to include the statistics of postfiltered outputs that are generated by a Gaussian filtering operation following the Gabor filter. The new method typically requires an order of magnitude less computation to design a filter than a previously proposed method. Experimental results demonstrate the efficacy of the method.

AB - Gabor filters have been successfully applied to a broad range of image processing tasks. The present paper considers the design of a single filter to segment a two-texture image. A new efficient algorithm for Gabor-filter design is presented, along with methods for estimating filter output statistics. The algorithm draws upon previous results that showed that the output of a Gabor-filtered texture is modeled well by a Rician distribution. A measure of the total output power is used to select the center frequency of the filter and is used to estimate the Rician statistics of the Gabor-filtered image. The method is further generalized to include the statistics of postfiltered outputs that are generated by a Gaussian filtering operation following the Gabor filter. The new method typically requires an order of magnitude less computation to design a filter than a previously proposed method. Experimental results demonstrate the efficacy of the method.

UR - http://www.scopus.com/inward/record.url?scp=0030410515&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030410515&partnerID=8YFLogxK

U2 - 10.1016/S0031-3203(96)00047-7

DO - 10.1016/S0031-3203(96)00047-7

M3 - Article

VL - 29

SP - 2005

EP - 2015

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 12

ER -