Search

The interaction matrix significantly affects the correction accuracy of wavefront correctors. For diffractive liquid crystal wavefront correctors (LCWFCs), the diffraction wavefront error must be taken into account to improve the accuracy of the interaction matrix. In this paper, a tunable Zernike polynomial coefficient method is demonstrated that decreases the effect of the diffraction wavefront error on the interaction matrix. Moreover, to eliminate the effect of the random error, a least squares method is combined with the tunable coefficient method. Experimental results show that, with the combined method, the error of the interaction matrix is decreased by a factor of 2.5 compared to that when the least squares method is used alone. Furthermore, an open loop adaptive correction experiment was performed and a considerable improvement of the image resolution was obtained by the novel method. Therefore, this method is very useful for liquid crystal adaptive optics systems to acquire the high correction accuracy. [ABSTRACT FROM AUTHOR]

Electron multiplying charge-coupled-device (EMCCD) applied in Shack–Hartmann wavefront sensor (S–H WFS) makes the wavefront sensing more efficient for adaptive optics (AO). However when the brightness of the observed target changes in large ranges in a few minutes, a fixed electron multiplying (EM) gain may not be optimum. Thus an auto-gain-control (AGC) method based on the spots image of the S–H WFS is proposed. The designed control value is the average value of the maximum signals of all the light spots in a frame. It has been demonstrated in the experiments that the control value is sensitive to the change of the target brightness, and is stable in the presence of detecting noises and turbulence influence. The goal value for control is predetermined based on the linear relation of the signal with the EM gain and the number of photons collected in sub-apertures. The conditions of the self-protection of the EMCCD are also considered for the goal value. Simulations and experiments indicate that the proposed control method is efficient, and keeps the sensing in a high SNR which reaches the upper SNR limit when sensing with EMCCD. The self-protection of the EMCCD is avoided during the whole sensing process. [ABSTRACT FROM AUTHOR]