The Shack-Hartmann wavefront sensor (SHWS) is an essential tool for wavefront sensing in adaptive optical microscopes. However, the distorted spots induced by the complex wavefront challenge its detection performance. Here, we propose a deep learning based wavefront detection method which combines point spread function image based Zernike coefficient estimation and wavefront stitching. Rather than using the centroid displacements of each micro-lens, this method first estimates the Zernike coefficients of local wavefront distribution over each micro-lens and then stitches the local wavefronts for reconstruction. The proposed method can offer low root mean square wavefront errors and high accuracy for complex wavefront detection, and has potential to be applied in adaptive optical microscopes.

Shack-Hartmann波前传感器（SHWS）是自适应光学显微镜中波前传感的重要工具。然而，由复杂波前引起的畸变点挑战了其检测性能。在此，我们提出了一种基于深度学习的波阵面检测方法，该方法将基于点扩散函数图像的Zernike系数估计和波阵面拼接相结合。该方法不是使用每个微透镜的质心位移，而是先估计每个微透镜上局部波阵面分布的Zernike系数，然后拼接局部波阵面以进行重建。所提出的方法可以提供低的均方根波阵面误差和高精度的复杂波阵面检测，并有潜力在自适应光学显微镜中应用。