Invariant pattern recognition using the RFM descriptor

A pattern descriptor invariant to rotation, scaling, translation (RST), and robust to additive noise is proposed by using the Radon, Fourier, and Mellin transforms. The Radon transform converts the RST transformations applied on a pattern into transformations in the radial and angular slices of the Radon transform data. These beneficial properties of the Radon transform make it an useful intermediate representation for the extraction of invariant features from patterns for the purpose of indexing/matching. In this paper, invariance to RST transformations is obtained by applying the 1D Fourier–Mellin and discrete Fourier transforms on the radial and angular slices of the Radon transform data respectively. The implementation of the proposed descriptor, which is based mainly on the fusion of the Radon and Fourier transforms and on a modification of the Mellin transform, is reasonably fast and correct. Theoretical arguments validate the robustness of the proposed descriptor to additive noise and empirical evidence on both occlusion/deformation and noisy datasets shows its effectiveness.