Identification of primary aberrations on a lateral-shearing interferogram using neural network

Abstract

A method is proposed for the identification of the primary aberrations with a single image of lateral-shear interferogram, obtained from a monochromatic laser source, using neural network. Neural network is used to teach the features of the some sampled interferograms and identify the primary aberrations of all the other interferograms by the nature of its excellence of interpolation and extrapolation. To increase the sensitivity of the fringe variation by primary aberrations, large amount of shear is given. Analysis of the pattern is performed to find proper features to teach the neural network with reduced time. Simulations are performed to verify the suggested method.