Features for texture segmentation using Gabor filters

Neena Mittal; D. P. Mital; Kap Luk Chan

doi:10.1049/cp:19990342

Features for texture segmentation using Gabor filters

Neena Mittal
, D. P. Mital
, Kap Luk Chan

Research output: Contribution to journal › Conference article › peer-review

Abstract

This work presents a method of extracting texture features from a Gabor transform data block and the application of these features for texture segmentation by clustering feature vectors. For a given image, 16 Gabor features using Gabor kernels with four scales and four orientations are computed. Filtered images are computed by using a bank of Gabor filter on a 32×32 windowed neighborhood for each pixel of the image. Texture features are obtained by computing the `energy' in the window for each pixel from the filtered images. Clustering algorithm is used to group the vectors based on their distribution in feature space. By clustering Gabor features, it is possible to segment an image into uniform regions. Experimental results demonstrate that features extracted using the proposed approach have excellent discriminating power.

Original language	American English
Pages (from-to)	353-357
Number of pages	5
Journal	IEE Conference Publication
Issue number	465 I
DOIs	https://doi.org/10.1049/cp:19990342
State	Published - 1999
Externally published	Yes
Event	Proceedings of the 1999 7th International Conference on Image Processing and its Applications - Manchester, UK Duration: Jul 13 1999 → Jul 15 1999

ASJC Scopus subject areas

Electrical and Electronic Engineering

Access to Document

10.1049/cp:19990342

Cite this

@article{2a27f20aa04b48cc982b7eea67376d01,

title = "Features for texture segmentation using Gabor filters",

abstract = "This work presents a method of extracting texture features from a Gabor transform data block and the application of these features for texture segmentation by clustering feature vectors. For a given image, 16 Gabor features using Gabor kernels with four scales and four orientations are computed. Filtered images are computed by using a bank of Gabor filter on a 32×32 windowed neighborhood for each pixel of the image. Texture features are obtained by computing the `energy' in the window for each pixel from the filtered images. Clustering algorithm is used to group the vectors based on their distribution in feature space. By clustering Gabor features, it is possible to segment an image into uniform regions. Experimental results demonstrate that features extracted using the proposed approach have excellent discriminating power.",

author = "Neena Mittal and Mital, \{D. P.\} and Chan, \{Kap Luk\}",

year = "1999",

doi = "10.1049/cp:19990342",

language = "American English",

pages = "353--357",

journal = "IEE Conference Publication",

issn = "0537-9989",

publisher = "Institution of Engineering and Technology",

number = "465 I",

note = "Proceedings of the 1999 7th International Conference on Image Processing and its Applications ; Conference date: 13-07-1999 Through 15-07-1999",

}

TY - JOUR

T1 - Features for texture segmentation using Gabor filters

AU - Mittal, Neena

AU - Mital, D. P.

AU - Chan, Kap Luk

PY - 1999

Y1 - 1999

N2 - This work presents a method of extracting texture features from a Gabor transform data block and the application of these features for texture segmentation by clustering feature vectors. For a given image, 16 Gabor features using Gabor kernels with four scales and four orientations are computed. Filtered images are computed by using a bank of Gabor filter on a 32×32 windowed neighborhood for each pixel of the image. Texture features are obtained by computing the `energy' in the window for each pixel from the filtered images. Clustering algorithm is used to group the vectors based on their distribution in feature space. By clustering Gabor features, it is possible to segment an image into uniform regions. Experimental results demonstrate that features extracted using the proposed approach have excellent discriminating power.

AB - This work presents a method of extracting texture features from a Gabor transform data block and the application of these features for texture segmentation by clustering feature vectors. For a given image, 16 Gabor features using Gabor kernels with four scales and four orientations are computed. Filtered images are computed by using a bank of Gabor filter on a 32×32 windowed neighborhood for each pixel of the image. Texture features are obtained by computing the `energy' in the window for each pixel from the filtered images. Clustering algorithm is used to group the vectors based on their distribution in feature space. By clustering Gabor features, it is possible to segment an image into uniform regions. Experimental results demonstrate that features extracted using the proposed approach have excellent discriminating power.

UR - https://www.scopus.com/pages/publications/0033365358

UR - https://www.scopus.com/pages/publications/0033365358#tab=citedBy

U2 - 10.1049/cp:19990342

DO - 10.1049/cp:19990342

M3 - Conference article

SN - 0537-9989

SP - 353

EP - 357

JO - IEE Conference Publication

JF - IEE Conference Publication

IS - 465 I

T2 - Proceedings of the 1999 7th International Conference on Image Processing and its Applications

Y2 - 13 July 1999 through 15 July 1999

ER -

Features for texture segmentation using Gabor filters

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this