Texture Image Classification Using Nonsubsampled Contourlet Transform and Local Directional Binary Patterns

Texture is a rich source of visual information about the surface characteristics of an object in the digital image. So texture characteristics play an important role in texture image classification. In this paper, we propose a novel approach of texture image classification based on nonsubsampled contourlet transform (NSCT) and local directional binary patterns (LDBP). The NSCT has translation invariability and LDBP has rotational invariance. The feature set is obtained by applying LDBP approach and co-occurrence parameters for three level NSCT subbands. The principal component analysis (PCA) is used to reduce the dimensionality of feature set. The class separability is enhanced using linear discriminant analysis (LDA). The features obtained from LDA are representatives of each texture class. The classification performance is tested on a set of 16 Brodatz textures. The k-NN classifier is used for classification. The experimental results demonstrate that the proposed method performs better than existing methods in the reduced feature set. Keywords— Nonsubsampled contourlet transform, Local directional binary pattern, Principal component analysis, Linear discriminant analysis, Texture.