Sliced orthogonal nonlinear generalized decomposition based fringe-adjusted joint transform correlation

A sliced orthogonal nonlinear generalized (SONG) decomposition based version of modified fringe-adjusted joint transform correlator (MFJTC), SBFJTC, is proposed. In SBFJTC scheme, the reference and input scenes are decomposed into a set of binary slices through SONG decomposition. MFJTC filtering of each pair of binary slices, corresponding to the same gray level of reference and input scene, is performed and then the outputs are added. Numerical simulations show that the proposed scheme delivers a better correlation performance and robustness to the white additive Gaussian noise, substitutive noise and nonoverlapping noise than that of both of the FJTC and the MFJTC.