Slope and curvature detection algorithm with high noise immunity in four-hole amplitude-modulated wavefront sensor

The four-hole amplitude-modulated wavefront sensor (FHAM-WS) can detect the slope and curvature of local wavefront in each subaperture using a four-hole mask modulated microlens array. In this paper, an efficient method is proposed to improve the precision and noise immunity of slope and curvature extraction algorithm in FHAM-WS. Firstly, preprocessing of the spot diagram for all subapertures acquired by FHAM-WS is conducted using a hybrid filter in both frequency and spatial domains. Then the spot image in each of the subapertures is labeled and divided into two parts: the central lobe and the side lobes. Slope of each local wavefront is detected using centroid estimation based on the central lobe. A hexagonal detection window is used to identify the central lobe and then the high moment algorithm is applied to estimate the centroids with the central lobe. Curvatures of each local wavefront are detected using intensity response expressions calculated from the side lobes. Simulations of the slope and curvature detection with ideal spot image under different noise conditions are conducted to verify the proposed method. The results demonstrate that the precision and noise immunity of the centroid estimation and intensity response for the proposed method are superior to those of the previous methods.