Evaluation and optimization of the Savitzky-Golay smoothing filter for noise reduction in thin film interference signal analysis

Petros I. Stavroulakis, Panos Liatsis, Nicholas Tipping, Paul Craddock

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

In this work, we present a comparison between Savitzky-Golay filtering and the traditional treatment of subtracting a dark signal to reduce noise from thin film interference spectra, acquired by a spectrometer in an adaptive system, where integration time needs to be varied for each measurement. We show that both classical Savitzky-Golay and dark signal removal are not optimal for interference spectra, however by applying a gradual variation of the Savitzky-Golay bin parameters across the signal, an adaptive version of the Savitsky-Golay technique can be used as a general-purpose noise reduction method for signals collected by an adaptive thin film thickness measurement system.

Original languageBritish English
Title of host publicationNovel Optical Systems Design and Optimization XVI
DOIs
StatePublished - 2013
Event16th Conference of Novel Optical Systems Design and Optimization - San Diego, CA, United States
Duration: 26 Aug 201328 Aug 2013

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume8842
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference16th Conference of Novel Optical Systems Design and Optimization
Country/TerritoryUnited States
CitySan Diego, CA
Period26/08/1328/08/13

Keywords

  • Adaptive bin size
  • Dark noise
  • Savitsky-Golay filtering
  • Signal to reconstruction error ratio
  • Thin film interferometry

Fingerprint

Dive into the research topics of 'Evaluation and optimization of the Savitzky-Golay smoothing filter for noise reduction in thin film interference signal analysis'. Together they form a unique fingerprint.

Cite this