Rotation-invariant texture features extraction using Dual-Tree Complex Wavelet Transform

Abstract

Rotation-invariant texture features extraction plays an important role in content based image retrieval. Texture features extraction based on wavelet transform are sensitive to texture rotation and translation. Thus, this paper proposes a new rotation invariant texture extraction technique using Principal Components Analysis (PCA) and Dual-Tree Complex Wavelet Transform (DT-CWT). Firstly, the angle of the principal direction of the texture image is calculated by the PCA. Then, the texture is rotated in the opposite direction by the same angle as detected by PCA. Finally, DT-CWT is applied to the preprocessed texture to extract features which are rotation invariant. Experiment proves the approximate shift invariance, good directional selectivity; computational efficiency properties of DT-CWT make it a good candidate for representing the rotation-invariant texture features.