Multispectra method for error reduction of the quantized Fourier modulus in the iterative Fourier transform algorithms used for phase retrieval

Abstract

A multispectra method is proposed for the error reduction of the quantized Fourier modulus in the iterative Fourier transform algorithms used for phase retrieval. The Fourier modulus is usually obtained by measuring the Fourier intensity with a digital device, thus the quantization error caused in the measurements of Fourier intensity leads to the error of the reconstructed object when iterative Fourier transform algorithms are used. The idea of the multispectra method uses several Fourier intensity distributions for various measuring ranges to generate a Fourier intensity distribution with low-quantization error. The error reduction of Fourier intensity with the multispectra method connects to the improvement of the phase retrieval with the iterative Fourier transform algorithms. Simulations are performed to demonstrate the effect of the multispectra method for phase retrieval with the iterative hybrid input-output algorithm.