Rotation and scaling invariant texture classification based on Gabor wavelets

Abstract

In this paper, an efficient texture classification method is proposed, which not only considers the effect of rotation, but also is scaling invariant. In our method, the Gabor wavelets are adopted to extract local features of an image, and the statistical property of the intensity values is used to represent the global feature. Then, an adaptive circular orientation normalization scheme is proposed for rotation normalization; and an elastic inter-frequency searching mechanism is used to reduce the effect of scaling. Our method is evaluated based on the Brodatz album, and the experimental results show that it outperforms the traditional algorithms.