Iteratively weighted centroiding for Shack-Hartmann wave-front sensors.

Several techniques have been used with Shack-Hartmann wavefront sensors to determine the local wave-front gradient across each lenslet. In this article we introduce an iterative weighted technique which is specifically targeted for open-loop applications such as aberrometers and metrology. In this article the iterative centroiding technique is compared to existing techniques such as center-of-mass with thresholding, weighted center-of-gravity, matched filter and cross-correlation. Under conditions of low signal-to-noise ratio, the iterative weighted centroiding algorithm is demonstrated to produce a lower variance in the reconstructed phase than existing techniques. The iteratively weighted algorithm was also compared in closed-loop and demonstrated to have the lowest error variance along with the weighted center-of-gravity, however, the iteratively weighted algorithm removes the bulk of the aberration in roughly half the iterations than the weighted center-of-gravity algorithm. This iterative weighted algorithm is also well suited to applications such as guiding on telescopes.