基于相关法的哈特曼图像网格自动划分方法

在哈特曼波前测量过程中,需要对哈特曼图像进行网格划分用于计算波前斜率。传统的网格划分方法通过观察哈特曼图像手动调整网格位置和网格间距。该方法不但需要人工干预,还给波前实时测量带来困难。为实现哈特曼图像网格划分的自动化,研究了基于相关法的哈特曼图像网格自动划分方法。该方法首先对哈特曼图像进行自相关求取网格行列数以及网格间距,然后利用该网格建立相应的模拟光斑图像,最后对模拟光斑图像和哈特曼图像进行互相关求取网格最佳位置。该方法无需人工干预,耗时短,实现了网格实时自动划分。

Automatic meshing of Shack-Hartmann image based on image correlation

In the process of Shack-Hartmann wave-front measure, meshing of Shack-Hartmann image is widely used for calculating wave-front slope. The conventional meshing method, which manually adjusts the mesh by observing the Shack-Hartmann image, is time-consuming. It is difficult to achieve real-time wave-front measure. In order to realize auto-meshing, in this paper, auto-meshing method based on image correlation is presented. In this algorithm, auto-correlation of Shack-Hartmann image is used for calculating grid interval and grid number. Then, simulative light-spot image according with the grid is obtained. At last, correlation operation between the simulative light-spot image and the Shack-Hartmann image is done to acquire optimal coordinate of the grid. This algorithm which is efficient and has no need for human intervention can achieve real-time auto-meshing.