Search

To measure the wave-front aberration, the Shack-Hartmann(SH) wave front sensor(WFS) is used in most adaptive optics(AO) systems. The accuracy of the centoid(Center of gravity) detecting of sub-images plays an important role in the performance of SHWFS. Normally the centriods of the point source spots are estimated to determine the local wavefront slopes by a center-of-gravity (CoG) algorithm. However the CoG algorithm suffers from several weaknesses. At first, there are few situations when a standard point source is available, such as situations with strong atmosphere disturbance and low flux. Secondly, CoG suffers significantly from CCD read-out noise. Instead of the traditional CoG, the correlation algorithm to estimate the wave-front slopes is used here, and the main characteristics of this method is analyzed. Analytical results predict this correlation approach presents better linear and robust performance. Moreover both simulation and experimental data from a point source based indoors experiment are carried out to verify the expected conclusions. The result agrees well with the simulation analysis.

A detailed analysis of the principle of Shack-Hartmann wavefront sensor is presented. Typical performance metrics such as dynamic range, sensitivity, repeatability and accuracy are defined and discussed. A method with which to calibrate a Shack-Hartmann wavefront sensor for absolute wavefront measurement for tilt without any accessorial apparatus is studied, and the method is tested by experiment.