Constrained optimization on the nonzero-order joint transform correlator constructed with the Mach–Zehnder configuration

Abstract

Although the implementation of a nonzero-order joint transform correlator by the Mach–Zehnder technique directly removes the zero-order component from the joint transform power spectrum in one step, the large sidelobe energy reduces the recognition ability. In this paper, a new method to improve the performance is achieved by using a constrained optimization technique to synthesize a reference function in the input spatial domain. The purpose is to reduce correlation sidelobes and to sharpen the correlation peak. Numerical results are presented to show the feasibility of the proposed technique.