Research And Application Of Sub-aperture Stitching Algorithm For Large Optical Elements Surface

With the constant development of national defense industry and advanced science and technology,large aperture planar optical elements play an increasingly significant role in astronomical telescopes,laser atmospheric transmission and other fields.During manufacturing and application of large aperture planar optical elements,inspection and tracking are carried out to determine whether the processing quality or operation status of the large aperture planar optical elements meet the engineering index.It is beneficial to adopt appropriate methods to maintain and compensate in time and reduce the costs.Sub-aperture stitching detection technology sacrifices part of the measurement efficiency to increase the extensibility of the transverse and longitudinal measurement range.Compared with traditional interferometry,Ricky-Common method and other detection technologies with large-scale measurement,it does not need to make large-aperture reference gauge,which reduces the measurement cost while retaining the measurement accuracy of small-aperture instruments.It can be seen that the sub-aperture stitching technology has a balance of accuracy,cost and efficiency in the detection of large-aperture optical elements,which has a good practical application value for its research.This project use sub-aperture stitching technology to detect the shape of large aperture optical elements,focusing on the study of the related algorithms involved in sub-aperture stitching.In this detection technique,the small-aperture Hartmann wavefront sensor is used as the detection instrument,and the corresponding subaperture data are collected according to the designed sub-aperture stitching scheme,and then the stitching results are obtained and analyzed by using the correlation algorithm.The thesis makes the following research:1.The paper respectively summarizes the existing slope reconstruction algorithm and subaperture stitching algorithm according their characteristics,and proposes the overlaying global slope error optimization sub-aperture stitching algorithm.In this algorithm,the least square optimization model is established based on the data information in the overlapping area of sub-aperture.After eliminating the tile-position error of each sub-aperture slope data,the weighted fusion is carried out to obtain the full aperture slope data.Finally,the full aperture surface is obtained by using the slope reconstruction algorithm.Simulation results show that the algorithm has good performance in high noise environment.The stitching accuracy is the same as the whole shape stitching accuracy,but better than the shape stitching in time efficiency.In the case of large number of sub-apertures,it is not easy to be affected by the error accumulation of local slope splicing,so the algorithm is feasible to some extent.2.The paper studies the principle of sparse sub-aperture stitching algorithm and proposes a sparse sub-aperture slope Zernike fitting stitching algorithm aiming at the limitation that the data objects processed by the existing algorithm are surface data.The algorithm uses the slope Zernike polynomial to establish the least square optimization model between the sparse subaperture slope data and the full-aperture slope data.After fitting the coefficient of the Zernike polynomial,the full-aperture surface can be fitted.The simulation results show that this algorithm has good anti-noise performance,and it has the same stitching accuracy as similar algorithms with the same number of sub-apertures and the same distribution,so it has certain application value.3.A sub-aperture stitching surface detection system based on Shack-Hartmann wavefront sensor is designed.A diameter of 100mm×150mm rectangular mirror is tested.And the experimental results are compared with the interferometer measurement results combined with the relevant evaluation indexes.The residual surface error in PV of the overlaying global slope error optimization of sub-aperture stitching algorithm is 0.005λ,and the residual error in RMS is 0.0018λ,the residual surface error in PV of the sparse sub-aperture slope Zernike fitting stitching algorithm is 0.0065λ,and the residual error in RMS is 0.0043λ(λ=532nm).It is further verified that the proposed algorithm is feasible and can be applied to the stitching system in which the collected sub-aperture is slope data.