Super-resolution imaging method and equal optical path and non-equal optical path super-resolution imaging system

Abstract

The present invention discloses a fractional Fourier transform domain-based super-resolution imaging method, a fractional Fourier transform domain-based equal-optical path super-resolution imaging system and a fractional Fourier transform domain-based non-equal-optical-path super-resolution imaging system. The super-resolution imaging method includes the following steps that: N intensity images of different fractional orders of an object are acquired; the complex amplitude of a target plane is initialized through using a random function, transformation of different fractional orders is carried out, so that the corresponding phase and amplitude of a measurement plane can be obtained, phases information is reserved, and amplitude information is discarded; N measured amplitude images and corresponding reserved phases are combined into complex amplitudes, and inverse fractional Fourier transform is performed on the complex amplitudes, so that corresponding N target plane images are obtained, the arithmetic mean values of the amplitudes and phases of the N target plane images are obtained, and the mean values of the amplitudes and phases are combined, and an obtained value is the update value of the complex amplitude of the target plane in next iteration; and the above steps are adopted to perform iterative calculation constantly, so that reconstructed complex amplitudes are obtained, and complex amplitudes of corresponding intensities are reconstructed images, and a super-optical resolution reconstructed image can be obtained through a plurality of times of iteration.

Description

technical field [0001] The invention relates to the field of optical technology, in particular to a super-resolution imaging method based on a fractional Fourier transform domain and an equal optical path and non-equal optical path super-resolution imaging system. Background technique [0002] With the development of computer technology, super-resolution image reconstruction algorithms have received more and more attention. Currently, Virtually Structured Detection (VSD) and Ptychographic Iterative Engine (PIE) have become two mainstream super-resolution methods. [0003] VSD can directly realize the illumination light modulation on the image detection device through the virtual grating modulation. By modulating the illumination light, the purpose of obtaining information in different frequency domains of the sample is achieved. After multiple measurements, the application of digital image synthesis technology breaks through the limitation of the traditional cut-off freque...

AI Technical Summary

Problems solved by technology

This type of method is essentially image post-processing, and the super-resolution effect is not obvious, and it does not solve the problem of super-resolution imaging from the perspective of the actual optical system. The processing effect for different images is also inconsistent, and the algorithm has low portability.

[0006] The PIE technology processes the far-field intensity sequence sequentially according to the idea of ​​​​executing in order of position, and uses the last result after processing as the initial value of the next calculation. The execution efficiency is relatively low, and there is a phenomenon of error accumulation in the serial calculation structure.

Images