Maximum-likelihood estimation in ptychography in the presence of Poisson-Gaussian noise statistics

Abstract

Optical measurements often exhibit mixed Poisson-Gaussian noise statistics, which hampers image quality, particularly under low signal-to-noise ratio (SNR) conditions. Computational imaging falls short in such situations when solely Poissonian noise statistics are assumed. In response to this challenge, we define a loss function that explicitly incorporates this mixed noise nature. By using maximum-likelihood estimation, we devise a practical method to account for camera readout noise in gradient-based ptychography optimization. Our results, based on both experimental and numerical data, demonstrate that this approach outperforms the conventional one, enabling enhanced image reconstruction quality under challenging noise conditions through a straightforward methodological adjustment.

arXiv commentContains main and supplementary documents

References in corpus (8)

• Experimental robustness of Fourier Ptychography phase retrieval algorithms
• Imaging With Nature: Compressive Imaging Using a Multiply Scattering Medium
• Using Automatic Differentiation as a General Framework for Ptychographic Reconstruction
• Full-field Fourier ptychography (FFP): spatially varying pupil modeling and its application for rapid field-dependent aberration metrology
• Three dimensions, two microscopes, one code: automatic differentiation for x-ray nanotomography beyond the depth of focus limit
• Efficient and flexible approach to ptychography using an optimization framework based on automatic differentiation
• The CRLB and Maximum likelihood in ptychography with Poisson noise
• Optimizing illumination for precise multi-parameter estimations in coherent diffractive imaging

Cited by in corpus (2)

• Bayesian multi-exposure image fusion for robust high dynamic range ptychography
• Noise-robust latent vector reconstruction in ptychography using deep generative models