Research of sub-aperture stitching algorithm for error correction in aspheric surface testing

In the testing of optical surfaces with large aperture or large aperture angle, sub-aperture stitching technology has been developed for a long time, while most testing analysis is for the rotationally symmetrical surface. In view of a cylindrical surface sub-aperture stitching algorithm based on discrete phase data, by using Legendre polynomials to fit the wave surface, numerical simulation and analysis for several main influencing factors have been carried out. For instance, a bicubic interpolation method has been used for non-numerical point in a discrete phase and fuzzy points in overlapping areas during the process of stitching sub-apertures, on account of its accuracy and calculation time. By considering alignment error, random noise, position error, especially rotation error in non-rotationally symmetrical cylindrical surface testing, we stimulate 9 sub-aperture phase data from a measured phase data to analyze the correction coefficient which determines the percentage of error correction in stitching algorithm. While the correction coefficient is approaching 1, the stitching result is most likely overcorrected. In order to achieve the best error correction, the correction coefficient should be set at around 0.6. Furthermore, combined with chaos particle swarm optimization, we reduce the possibility of the algorithm falling into the local optimum, and improve the accuracy of stitching.