Modified joint transform correlator binarized by error diffusion II Spatially variant range limit

Nonlinear scaling of the joint power spectrum in one approach and noise-subtraction followed by nonlinear scaling in a second approach are analyzed in conjunction with error-diffusion binarization for joint transform correlator configurations in the presence of input-additive white Gaussian noise. Nonlinear scaling is performed with a spatially variant range limit. For the second approach the subtracted noise value is an estimate of the noise power spectral density. Computer simulations and optical experiments demonstrate the advantages of the proposed error-diffusion-based joint transform correlator approaches. In particular the approach based on noise subtraction before nonlinear scaling is advantageous even for very high noise levels.