Diffusion Approach for Texture Analysis Based on Ldbp

Texture classification plays an important role in computer vision and image processing applications, which has received considerable attention over the last several decades. In this paper, diffusion approach for texture analysis based on anisotropic diffusion and local directional binary patterns (LDBP) is presented. The proposed approach uses partial differential equation (PDE) as the pre processing step to obtain texture component of the image. The feature set is obtained by applying LDBP approach to texture component and then extracting co-occurrence parameters. The separability of texture feature classes is enhanced using linear discriminant analysis (LDA). The features obtained from LDA are class representatives. The proposed approach is validated on four texture datasets: Brodatz, Vistex, Kylberg and Oulu. The discriminating power of the proposed method is evaluated using tenfold validation of k-NN classifier. The experimental results indicate that the proposed approach performs better than the other methods in the literature.