Abstract:Phase retrieval (PR) technology can reconstruct the complex amplitude information of the entrance pupil surface from the captured defocus intensity images, and has been successfully applied in many scientific fields such as astronomical observation, biomedical imaging and digital signal restoration. For low numerical aperture (NA) optical systems, the point spread function (PSF) can be calculated using the Fresnel diffraction propagation theory based on Fourier transform, ignoring the polarization field of the light field. However, for high-NA systems, the focal spot not only includes the transverse component, but also the longitudinal component accounts for a proportion that cannot be ignored, thus necessitating a vector model when calculating the PSF in the non-paraxial region. In this paper, the vector diffraction calculation model with arbitrary defocus distance based on the extended Nijboer-Zernike (ENZ) theory is established, which can characterize the light field components in three directions under the Cartesian coordinate system. What's more, a modified Gauss-Newton theory is innovatively applied to the axial phase difference PR model. Compared with the traditional first-order algorithm, the second-order algorithm can reduce the number of iterations. In addition, so as to verify the effectiveness of the proposed method, numerical simulations matching the physical model are carried out, the results show that the proposed method can accurately reconstruction the wavefront with high robustness. In conclusion, the established vector PR model in this paper would provide a creatively guidance for wavefront measurement of high-NA optical systems that is significant in the fields of semiconductor lithography, micro imaging and micro manipulation.

Exit Wave Function Reconstruction from Two Defocus Images Using Neural Network.

Exit-wave Phase Retrieval from a Single High-Resolution Transmission Electron Microscopy Image of a Weak-Phase Object

Alleviating Image Artifacts in Wavefront Coding Extended Depth of Field Imaging System

Wavefront Reconstruction Based on Deep Transfer Learning for Microscopy.

Exit wave reconstruction of a focal series of images with structural changes in high‐resolution transmission electron microscopy

Aberration Correction Method For Fourier Ptychographic Microscopy Based On Neural Network

Modal-based Adaptive Autofocusing Phase Retrieval for Wavefront Measurement

Application of Vector Phase Retrieval Model in High Numerical Aperture Optical Systems

Extended Depth of Field Reconstruction with Complex Field Estimation

Exit Wavefunction Reconstruction from Single Transmission Electron Micrographs with Deep Learning

A New Scheme for the Exit-Wave Reconstruction from A Small Set of Focus Series of Hrem Images

REW– Exit‐wave Reconstruction and Alignments for Focus‐variation High‐resolution Transmission Electron Microscopy Images

Real-time phase-retrieval and wavefront sensing enabled by an artificial neural network

Imaging Simulation of Charged Nanowires in TEM with Large Defocus Distance.

A New Method for Axial Decay Function Calibration of Evanescent Field in Multi-Angle Total Internal Reflection Fluorescence Microscopy

A Deep-Learning-based Workflow to Deal with the Defocusing Problem in High-Throughput Experiments

NeRFocus: Neural Radiance Field for 3D Synthetic Defocus

Reconstruction of nanoscale particles from single-shot wide-angle FEL diffractions patterns with physics-informed neural networks

Super-Resolution Image Reconstruction of Wavefront Coding Imaging System Based on Deep Learning Network

Three-dimensional virtual refocusing of fluorescence microscopy images using deep learning

Reconstruction Of 3-D Defect Profiles From Mfl Signals Using Radial Wavelet Basis Function Neural Network