OSA's Digital Library

Chinese Optics Letters

Chinese Optics Letters

| PUBLISHED MONTHLY BY CHINESE LASER PRESS AND DISTRIBUTED BY OSA

  • Vol. 4, Iss. 4 — Apr. 1, 2006
  • pp: 243–246

Mixture gas component concentration analysis based on support vector machine and infrared spectrum

Peng Bai and Junhua Liu  »View Author Affiliations


Chinese Optics Letters, Vol. 4, Issue 4, pp. 243-246 (2006)


View Full Text Article

Acrobat PDF (413 KB)





Browse Journals / Lookup Meetings

Browse by Journal and Year


   


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Article Tools

Share
Citations
  • Export Citation/Save Click for help

Abstract

A novel quantitative analysis method of multi-component mixture gas concentration based on support vector machine (SVM) and spectroscopy is proposed. Through transformation of the kernel function, the seriously overlapped and nonlinear spectrum data are transformed in high-dimensional space, but the high-dimensional data can be processed in the original space. Some factors, such as kernel function, range of the wavelength, and penalty coefficient, are discussed. This method is applied to the quantitative analysis of natural gas components concentration, and the component concentration maximal deviation is 2.28%.

© 2005 Chinese Optics Letters

OCIS Codes
(070.4790) Fourier optics and signal processing : Spectrum analysis
(200.4740) Optics in computing : Optical processing
(300.6340) Spectroscopy : Spectroscopy, infrared

Citation
Peng Bai and Junhua Liu, "Mixture gas component concentration analysis based on support vector machine and infrared spectrum," Chin. Opt. Lett. 4, 243-246 (2006)
http://www.opticsinfobase.org/col/abstract.cfm?URI=col-4-4-243

You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Log in to access OSA Member Subscription

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Log in to access OSA Member Subscription

« Previous Article  |  Next Article »

OSA is a member of CrossRef.

CrossCheck Deposited