Finite-dimension eigen frequency domain analysis method for optical imaging system

Abstract

The invention discloses a finite-dimension eigen frequency domain analysis method for an optical imaging system. The finite-dimension eigen frequency domain analysis method comprises the following steps: a first step: solving a two-dimensional light intensity transmission matrix of an optical system according to a PSF matrix and a one-dimensional object vector of the optical system; a second step: solving a characteristic value vector of the two-dimensional light intensity transmission matrix by using a QR decomposition method, and solving an eigen function vector group of the two-dimensional light intensity transmission matrix by using a power method; and a third step: solving object and image finite-dimension eigen spectral vectors by means of the object vector, an image vector and the eigen function vector group. The finite-dimension eigen frequency domain analysis method for the optical imaging system is used for overcoming the defects of the Fourier optical frequency-domain imaging analysis method in theory and numeral value calculation, providing the solving method of the object and image eigen spectral vectors and realizing that the object eigen spectral vector multiplying by the characteristic value vector of the optical imaging system is equal to the image eigen spectral vector.

Description

technical field [0001] The invention belongs to the technical field of optical system information optics and imaging analysis, and relates to a finite-dimensional intrinsic frequency domain analysis method for an optical imaging system. Background technique [0002] According to the analysis method of Fourier optics, the convolution process of optical system imaging spatial domain imaging can be described by the spectral product relationship in the Fourier frequency domain. The basic principle of the Fourier optical frequency domain imaging analysis method is to perform Fourier transform on the point spread function (PSF) of the object, image and optical system to obtain the object spectrum, image spectrum and optical transfer function (OTF) of the optical system respectively. The image spectrum is equal to the product of the object spectrum and the OTF. However, the frequency-domain imaging analysis method of Fourier optics has imperfections in theory and numerical calcula...

AI Technical Summary

Problems solved by technology

[0004] In terms of numerical calculation, since the integral transformation kernel of Fourier transform is a complex exponential function (combination of cosine function and sine function), and complex exponential function is an unbounded function with infinite data volume, it is impossible to use a computer for numerical calculation

At present, the fast Fourier transform method (FFT) is generally used to realize the numerical calculation of the theoretical formula of Fourier transform, but the calculation result of FFT is inconsistent with the calculation result of the Fourier transform formula, so it is impossible to obtain accurate numerical calculation results.

Images