Email Document Reference

Enter your email address below and the reference for this document will be sent to shortly from webmaster@ceesi.com.

Title: Outlier-Resistant Fitting Of Gray-Scale Images Illustrated By Optical Fiber Geometry
Author: Dominic Vecchia, Chill-Ming Wang, Matt Young
Source: 1993 Measurement Science Conference
Year Published: 1993
Abstract: Using the measurement of geometric parameters of optical fiber ends as an example, we discuss a curve-fitting technique, called least median of squares (LMS), that is highly resistant to the influence of outlying data points. The LMS method is a form of robust regression that can resist the effect of nearly 50% contaminated data. W C fit ellipses to the digitized gray-scale images of each of two fiber ends by LMS and by ordinary leastsum- of-squares regression. The two methods yield nearly identical results on a pristine fiber end, but the LMS method is far superior on a damaged fiber end, even though we make no effort to identify spurious data points explicitly.

In order to prevent spam and automated file downloads for documents within the Measurement Library, please follow the instructions below and then you will be able to email a reference to this article.