Paper
1 January 1987 Regression Analysis Of Zernike Polynomials
Louis D. Grey
Author Affiliations +
Abstract
In evaluating the performance of a mirror by interferometric methods, a standard procedure used by the optical engineer is to obtain the optical path difference (OPD) map from the fringes of the interferogram and to express the differences in terms of a least squares fit of the classical Zernike polynomials. Since the Zernike polynomials represent the aberrations inherent in the optics, the engineer can determine whether or not the optics will meet performance requirements and if not, what measures need to be taken to reduce the aberrations. This paper examines the least squares fitting process from a statistical point of view and points out the necessary steps and precautions which must be taken to ensure that the analysis rests on a firm foundation so that the conclusions drawn by the optical engineer are warranted. The fitting process is examined from the standpoint of statistical linear regression theory and the following concepts and ideas are introduced: 1) How to determine whether the fit is a good fit and what to do about outliers. 2) Whether the underlying assumptions of normality and constant variance have been met, and, if not, what should be done. 3) How to compute confidence intervals 4) What do the correlation and covariance matrices tell us. 5) How to look at the information contained in the residuals and how to detect influential observations. Regression analysis has proved to be a powerful tool in other fields and there is every reason to believe it can make a significant contribution to optical engineering.
© (1987) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Louis D. Grey "Regression Analysis Of Zernike Polynomials", Proc. SPIE 0818, Current Developments in Optical Engineering II, (1 January 1987); https://doi.org/10.1117/12.978912
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Zernike polynomials

Statistical analysis

Electro-optical engineering

Error analysis

Optical engineering

Antimony

Distance measurement

Back to Top