计算扩展目标夏克-哈特曼波前传感器子图像偏移量的相关函数质心法研究

发布时间：2018-10-24 07:30

【摘要】：扩展目标夏克-哈特曼波前传感器子图像之间偏移量的计算是影响波前传感精度的关键,通常采用相关算法来实现,并通过抛物线插值等方法来达到亚像元精度。子图像相对偏移量的计算也可以采用计算相关函数质心的方法来实现,其主要步骤是先计算子图像间的相关函数,在此基础上计算相关函数的质心,达到亚像元精度。通过仿真研究表明这种算法的精度与进行相关运算时的图像大小、计算质心时的相关函数窗口大小以及相关函数阈值的设定有关;同时,图像的信噪比也会影响算法的精度。研究表明,图像的信噪比小于1时,质心算法的计算误差相对较大;当图像的信噪比高于2时,相关函数质心算法的误差大约是抛物线插值法误差的一半。实验结果与仿真结果也基本吻合。
[Abstract]:The calculation of the offset between the sub-images of the extended Shacker-Hartmann wavefront sensor is the key to affect the accuracy of the wavefront sensor. The correlation algorithm is usually used to achieve the sub-pixel accuracy and parabola interpolation is used to achieve the sub-pixel accuracy. The calculation of relative offset of sub-image can also be realized by calculating the centroid of correlation function. The main step is to calculate the correlation function between sub-images first and then calculate the center of mass of correlation function on this basis to achieve the accuracy of sub-pixel. The simulation results show that the accuracy of the algorithm is related to the size of the image, the window size of the correlation function and the threshold of the correlation function when the center of mass is calculated, and the SNR of the image will also affect the accuracy of the algorithm. The results show that when the SNR of the image is less than 1, the calculation error of the centroid algorithm is relatively large, and when the SNR of the image is higher than 2, the error of the correlation function centroid algorithm is about half the error of the parabola interpolation method. The experimental results are in good agreement with the simulation results.
【作者单位】：北京理工大学光电学院;
【基金】：国家973计划(2014CB340004)
【分类号】：TP212

【相似文献】