Automatic adjustment method of computer-aided interferometer based on deep learning

Abstract

The invention discloses an automatic adjustment method of a computer-aided interferometer based on deep learning. The method comprises steps of collecting an interferogram and three-dimensional coordinates thereof as training data; classifying the expanded training data based on a K-means clustering algorithm to obtain a data set; dividing the data set into a training set and a verification set byadopting a k-fold cross validation method, and training the VGG-16 network model by utilizing the training set and the verification set obtained by each fold; and inputting an interferogram acquiredin real time into the trained VGG-16 network model to obtain the misalignment amount of the interferometer. According to the method, the specific interference pattern deviation value does not need tobe solved; trained network classification data are fully utilized, the coordinates of the position where the interferometer is located are determined by classifying interferograms formed by the interferometer, then the mirror to be measured is controlled to move to the accurate position, the advantages of the deep convolutional network are brought into full play, and the method has the advantagesof being good in robustness and high in accuracy.

Description

technical field [0001] The invention belongs to the technical field of image processing and computer vision, and specifically relates to an automatic adjustment method of a computer-aided interferometer based on deep learning. Background technique [0002] Computer Aided Alignment (Computer Aided Alignment, CAA) refers to a new technology that breaks away from traditional alignment and mainly relies on human manual operation and operating experience, with the help of computers and precision mobile equipment. In the optical field, various precise and complex The equipment and devices are widely used, relying on the traditional adjustment method has low accuracy and takes a long time. Using computer-aided adjustment technology can solve the positional misalignment of the optical system in real time, and then control the mechanical structure to adjust the position of the optical components. Computer-aided assembly technology makes the optical system organically combined with el...

AI Technical Summary

Problems solved by technology

This method improves the fitting accuracy of the Zernike polynomial, but does not fundamentally solve the error of the computer simulation model

Images