Classification of Rotated and Scaled Textured Images Using Spectral Moments

This paper describes a classification method for rotated and scaled textured images using invariant parameters based on spectral-moments. Although it is well known that rotation invariants can be derived from moments of grey-level images, the use is limited to binary images because of its computational unstableness. In order to obtain computationally stable moments, we use power spectrum instead of the grey levels to compute moments and adjust an integral region to scale change. Rotation and scale invariants are obtained as the ratio of the rotation invariants on the basis of a spectral-moment property with respect to scale. The effectiveness of the approach is illustrated through experiments on natural textures from the Brodatz album. The stability of the invariants with respect to change of scale is discussed theoretically and confirmed experimentally.