Enhanced design of multiplexed coded masks for Fresnel incoherent correlation holography

• Published in: Scientific reports Vol. 13; no. 1; p. 7390
• Main Authors: Gopinath, Shivasubramanian, Bleahu, Andrei, Kahro, Tauno, John Francis Rajeswary, Aravind Simon, Kumar, Ravi, Kukli, Kaupo, Tamm, Aile, Rosen, Joseph, Anand, Vijayakumar
• Format: Journal Article
• Language: English
• Published: England Nature Publishing Group 06.05.2023; Nature Publishing Group UK; Nature Portfolio
• Subjects: Algorithms; Holography; Light; Polarization; Sensors; Simulation
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-023-34492-2

Fresnel incoherent correlation holography (FINCH) is a well-established incoherent digital holography technique. In FINCH, light from an object point splits into two, differently modulated using two diffractive lenses with different focal distances and interfered to form a self-interference hologram. The hologram numerically back propagates to reconstruct the image of the object at different depths. FINCH, in the inline configuration, requires at least three camera shots with different phase shifts between the two interfering beams followed by superposition to obtain a complex hologram that can be used to reconstruct an object's image without the twin image and bias terms. In general, FINCH is implemented using an active device, such as a spatial light modulator, to display the diffractive lenses. The first version of FINCH used a phase mask generated by random multiplexing of two diffractive lenses, which resulted in high reconstruction noise. Therefore, a polarization multiplexing method was later developed to suppress the reconstruction noise at the expense of some power loss. In this study, a novel computational algorithm based on the Gerchberg-Saxton algorithm (GSA) called transport of amplitude into phase (TAP-GSA) was developed for FINCH to design multiplexed phase masks with high light throughput and low reconstruction noise. The simulation and optical experiments demonstrate a power efficiency improvement of ~ 150 and ~ 200% in the new method in comparison to random multiplexing and polarization multiplexing, respectively. The SNR of the proposed method is better than that of random multiplexing in all tested cases but lower than that of the polarization multiplexing method.