High speed aberration correction method based on machine learning

Abstract

The invention discloses a high speed aberration correction method based on machine learning. The incident parallel beam passes through the spatial light modulator that does not load the wavefront phase distribution to obtain the ideal focused spot. A series of wavefront phase distributions are obtained using the Zenike polynomial processing, and each wavefront phase distribution is loaded into the spatial light modulator to obtain a distorted focused spot. The light intensity distribution of each distorted focused spot and the respective Zenike coefficients under the incident wave wavefront distribution are input to the machine learning training to obtain the correction model. The intensity distribution of the distorted focused spot pattern of the scattering medium to be measured is input to the calibration model to obtain the values of the respective Zenike coefficients, and by taking the negative value calculation, the corrected phase distribution is obtained and loaded into the spatial light modulator to achieve aberration correction. The method can realize the high speed optical aberration correction of the optical path, the correction speed is fast and the accuracy is high, and the problem that the traditional adaptive optical algorithm is slow is solved.

Description

technical field [0001] The invention belongs to the field of optical microscopic imaging, in particular to a high-speed aberration correction method based on machine learning. Background technique [0002] When microscopic imaging is used for deep imaging of biological tissue, the wavefront aberration caused by the production accuracy error of optical components and the inhomogeneity of the refractive index of biological tissue in the optical path will seriously affect the focusing of the spot, and with the imaging depth The increase of the signal-to-noise ratio and resolution decreases, and the imaging quality drops sharply. In response to this phenomenon, researchers have proposed many solutions, among which adaptive optics technology is a common and effective solution. [0003] The principle of adaptive optics technology to improve imaging quality is as follows: First, the wavefront sensor is used to measure the wavefront distortion of the optical system caused by variou...

AI Technical Summary

Problems solved by technology

The weakness of the traditional adaptive correction algorithm is that the correction speed is very slow. Researchers have also proposed a series of algorithms to speed up the correction speed, such as the COAT algorithm

Images