Nonlinear joint transform correlators

摘要

A joint power spectrum based processor called joint transform correlator is presented. The joint power spectrum, which is the intensity of the Fourier transform of the input scene displayed alongside the reference function to be detected, contains the cross-power spectrum of the reference function and the input scene. The correlation functions are obtained by applying an inverse Fourier transform to the joint power spectrum. Nonlinear transformations of the joint power spectrum modify the Fourier magnitudes of the reference function and the input scene while preserving their Fourier phases. By adjusting the nonlinear transformations, correlation like functions with a variety of characteristics can be generated in the output plane. This nonlinear processor is shift invariant and can produce at the output well-defined peaks identifying the presence and the location of the targets.