Reconstruction of a complex-valued object in coherent imaging through a random-phase screen

Area of Science:

• Optics
• Image Reconstruction
• Phase Retrieval

Background:

• Coherent imaging through random-phase screens presents significant challenges for object reconstruction.
• Traditional phase retrieval methods often struggle with noise and inherent ambiguities.

Purpose of the Study:

• To develop and validate a novel phase retrieval method for complex-valued object reconstruction.
• To assess the impact of noise and random-phase screen characteristics on reconstruction accuracy.

Main Methods:

• Utilizing two intensity measurements of the object's Fourier transform.
• Employing exponential modulation of the image field to introduce phase dependence in the second measurement.
• Computer simulations to evaluate reconstruction robustness and error sources.

Main Results:

• The proposed method successfully reconstructs complex-valued objects, with a minor ambiguity of displacement.
• Reconstruction accuracy is robust against noise from finite exposure times and quantum effects.
• Reconstruction error correlates with the variance and correlation length of the random-phase screen.

Conclusions:

• The two-measurement phase retrieval technique offers a viable solution for imaging through random-phase screens.
• Understanding the influence of noise and screen properties is crucial for optimizing reconstruction fidelity.
• The method demonstrates practical applicability in scenarios requiring complex object imaging.