Far-field diffraction computational imaging based on parameter-robust illumination and direct phase optimization

TL;DR

This paper introduces a parameter-robust illumination scheme and a direct phase optimization algorithm to enhance far-field diffraction imaging, enabling accurate reconstruction of complex objects from noisy measurements without prior information.

Contribution

The work presents a novel parameter-robust illumination pattern and a direct phase optimization method to improve phase retrieval in coherent diffraction imaging under noisy conditions.

Findings

Effective reconstruction of complex objects from noisy diffraction data.

Enhanced phase retrieval performance demonstrated experimentally.

Robustness of the method in practical noisy measurement scenarios.

Abstract

Coherent diffraction imaging (CDI) is a promising imaging technique revealing most of the information from diffraction measurements. An ideal CDI should reconstruct complex-valued object from a single-shot far-field diffraction without any priori information about the target. To realize the ideal CDI, we propose a class of parameter-robust illumination pattern. A direct phase optimizing algorithm is also raised here to improve the performance of phase retrieval in strong noise. Experimental result demonstrates the efficiency of our scheme in practical noisy measurement for complex-valued target.