Implicitly Learned Neural Phase Functions for Basis-Free Point Spread Function Engineering

TL;DR

This paper introduces a neural network-based method for PSF engineering that overcomes traditional limitations, enabling more accurate and generalizable phase function design for advanced imaging applications.

Contribution

It proposes an implicit neural representation approach for PSF engineering, improving over pixel-wise optimization in accuracy and generalization.

Findings

Achieves higher median MSSIM (0.8162) than pixel-wise methods (0.0)

Attains higher median PSNR (10.38 dB) compared to pixel-wise optimization (6.653 dB)

Demonstrates improved generalization across diverse PSFs

Abstract

Point spread function (PSF) engineering is vital for precisely controlling the focus of light in computational imaging, with applications in neural imaging, fluorescence microscopy, and biophotonics. The PSF is derived from the magnitude of the Fourier transform of a phase function, making the construction of the phase function given the PSF (PSF engineering) an ill-posed inverse problem. Traditional PSF engineering methods rely on physical basis functions, limiting their ability to generalize across the range of PSFs required for imaging tasks. We introduce a novel approach leveraging implicit neural representations that overcome the limitations of pixel-wise optimization methods. Our approach achieves a median MSSIM of 0.8162 and a mean MSSIM of 0.5634, compared to a median MSSIM of 0.0 and a mean MSSIM of 0.1841 with pixel-wise optimization when learning randomly generated phase functions. Our approach also achieves a median PSNR of 10.38 dB and a mean PSNR of 8.672 dB, compared to a median PSNR of 6.653 dB and a mean PSNR of 6.660 dB with pixel-wise optimization for this task.