Noise removal in Shack-hartamnn wavefront sensor based on nonconvex weighted adaptively regularization

Shack-Hartmann wavefront sensor (SHWFS) is widely used to measure the wavefront aberrations in adaptive optics system. However, a conventional SHWFS has finite accuracy depending on the noises. The removal methods of additive noise in SH WFS have been analyzed. We put forward a new variation model to solve this problem. In this model, a nonconvex weighted regularization is used in which the regularization parameters can change adaptively according to the noise level. We present the advantages of this model in detail. The split Bregman algorithm and the augmented Lagrangian duality algorithm are used to resolve this nonconvex problem. The experiments results show that the subaperture focal plane spots are well preserved. The visual and quantitative evaluation including the peak signal to noise ratio (PSNR) and the centroid detecting error are also shown the effective of our method in removing the additive noise in the adaptive optics system.