Unsupervised texture segmentation of images based on local energy of Gabor features

Punya, Thitimajshima; Yuttapong, Rangsanseri; Prapon, Rakprathanporn; Tawee, Kusumawalee

Unsupervised texture segmentation of images based on local energy of Gabor features

Author(s)

Punya, Thitimajshima

Yuttapong, Rangsanseri

Prapon, Rakprathanporn

Tawee, Kusumawalee

Date Issued

December 1, 2002

Type

Conference Paper

Abstract

In this paper, we propose an unsupervised texture segmentation of images by clustering of feature vectors derived from local energy Gabor-filtered images. A cosine modulated Gabor filters with different radial frequencies and orientations were used. The algorithm was tested on many texture images and two examples are given to illustrate the effectiveness of such an algorithm.

Citation

Proceeding of the International Conference on Telecommunications, 1, 144-146, 2002