Complex wavelet co-occurrence based on thresholding for texture classification

The complex wavelet transform is superior to discrete wavelet transform in two aspects. Those are its shift invariant property and good directional selectivity. In this paper, we propose a new approach of combining dual-tree complex wavelet transform (DT-CWT) with traditional co-occurrence method for texture classification. Thresholding is a method to preserve the significant data of the image while discarding the insignificant part. On this basis for further enhancement of classification, we have acquired the texture images by applying thresholding technique to the entire texture database. These thresholded images are then applied to the fusion model of DTCWT with co-occurrence method. Results show that our proposed methods provide better classification accuracy than classical co-occurrence method.