Research On Subaperture Stitching Testing Method

Sub-aperture stitching testing method is to test different parts of an optical surface respectively and then stitch them together to get a whole testing result of the optical surface. It can expand the lateral and longitudinal dynamic range of the interferometer, so it can be used in testing large or aspheric optical surface. However, it can lower the measuring precision. Apparently, one way to improve the measuring precision is to improve the mechanical localization accuracy. However, the mechanical localization accuracy will greatly increase the cost of using the sub-aperture stitching testing method. Therefore, it is the problem faced by the sub-aperture stitching testing method how to improve the accuracy and quality of the sub-aperture stitching testing method within limited mechanical localization accuracy.In order to solve this problem, this dissertation focuses on designing testing system which is insensitive to mechanical localization errors, globally optimal algorithms which are used for compensating mechanical localization errors, and sub-aperture overlapping fusion algorithm which can preserve high-frequency components of the measurements. The major contents of the dissertation are as follows:An annular sub-aperture stitching method based on auto-collimation is proposed.In this method, we move a ball instead of the interferometer and the aspheric surface so that testing results for adjacent annular sub-apertures are registered. Thus, this method can stitch together the sub-aperture testing results under large mechanical localization errors.(2) A sub-aperture overlapping area fusion algorithm based on wavelet transformation is proposed. In this algorithm, high-frequency components and low-frequency components are fused by different rules. The high-frequency components are fused using the choose-randomly scheme, so this algorithm can retain the high-frequency components.(3) A sub-aperture stitching algorithm using image sharpening is proposed. In this algorithm, image sharpening is used to strengthen the high-frequency components of measurements. Thus, it can improve the stitching accuracy under small mechanical localization errors.(4) A sub-aperture stitching algorithm based on genetic algorithms is proposed. Under large mechanical localization errors, sub-aperture stitching algorithms based on the least squares method easily become trapped by locally optimal solutions. However, genetic algorithms are less likely to become trapped by locally optimal solutions, thus we utilize this method to develop a more robust sub-aperture stitching algorithm. With the help of genetic algorithms, this new sub-aperture stitching algorithm performs better than the algorithm based on the least squares method by increasing the possibility of finding the globally optimal solution.