Translation and scale invariance of 2D and 3D Hahn moments

摘要

Moment invariants of 3D point distributions may be computed that are ST(scale and translation) invariant, in the same way as 2D moment invariants. Hahn moment is a discrete orthogonal moment which can be a useful set of image feature descriptor. Their higher order moments have better image representation power as compared to legendre and Zernike moments. Hahn moments can be utilized to adaptively extract the global and local information of an image. 3D Hahn moments and their invariants can be used for image analysis and recognition of different 3D images, such as MRI or PET. In this paper, we have introduced TS(translation and scale) invariants of two and three-dimensional Hahn moments. The Experimental results are given in order to consolidate the theoretical derivations.