Share Email Print
cover

Proceedings Paper

Novel filter design algorithm for multivariate optical computing
Author(s): Olusola O. Soyemi; Paul J. Gemperline; Lixia Zhang; DeLyle Eastwood; Hong Li; Michael L. Myrick
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

A new algorithm for the design of optical computing filters for chemical analysis otherwise known as Multivariate Optical Elements (MOEs), is described. The approach is based on the nonlinear correlation of the MOE layer thicknesses to the standard error in sample prediction for the chemical species of interest using a modified version ofthe Gauss-Newton nonlinear optimization algorithm. The design algorithm can either be initialized by random layer thicknesses or by a pre-existing design. The algorithm has been successfully tested by using it to design a MOE for the determination of copper uroporphynn in a quaternary mixture of uroporphyrin (freebase), nickel uroporphyrin, copper uroporphynn, and tin uroporphyrin.

Paper Details

Date Published: 26 February 2001
PDF: 12 pages
Proc. SPIE 4205, Advanced Environmental and Chemical Sensing Technology, (26 February 2001); doi: 10.1117/12.417462
Show Author Affiliations
Olusola O. Soyemi, Univ. of South Carolina (United States)
Paul J. Gemperline, East Carolina Univ. (United States)
Lixia Zhang, Univ. of South Carolina (United States)
DeLyle Eastwood, Univ. of South Carolina (United States)
Hong Li, Univ. of South Carolina (United States)
Michael L. Myrick, Univ. of South Carolina (United States)


Published in SPIE Proceedings Vol. 4205:
Advanced Environmental and Chemical Sensing Technology
Tuan Vo-Dinh; Stephanus Buettgenbach, Editor(s)

© SPIE. Terms of Use
Back to Top