Independent component analysis for the identification of sources of variation on an industrial nirs application

Moya Gonzalez, Adolfo and Barreiro Elorza, Pilar and Jouan-Rimbaud Bouveresse, D. and Rutledge, Douglas N. (2011). Independent component analysis for the identification of sources of variation on an industrial nirs application. In: "Chimiométrie 2011", 01/12/2011 - 02/12/2011, Marsella, Francia. pp. 97-100.

Description

Title: Independent component analysis for the identification of sources of variation on an industrial nirs application
Author/s:
  • Moya Gonzalez, Adolfo
  • Barreiro Elorza, Pilar
  • Jouan-Rimbaud Bouveresse, D.
  • Rutledge, Douglas N.
Item Type: Presentation at Congress or Conference (Article)
Event Title: Chimiométrie 2011
Event Dates: 01/12/2011 - 02/12/2011
Event Location: Marsella, Francia
Title of Book: Proceedings of Chimiométrie 2011
Date: 2011
Subjects:
Faculty: E.T.S.I. Agrónomos (UPM) [antigua denominación]
Department: Ingeniería Rural [hasta 2014]
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[img]
Preview
PDF - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (514kB) | Preview

Abstract

A Near Infrared Spectroscopy (NIRS) industrial application was developed by the LPF-Tagralia team, and transferred to a Spanish dehydrator company (Agrotécnica Extremeña S.L.) for the classification of dehydrator onion bulbs for breeding purposes. The automated operation of the system has allowed the classification of more than one million onion bulbs during seasons 2004 to 2008 (Table 1). The performance achieved by the original model (R2=0,65; SEC=2,28ºBrix) was enough for qualitative classification thanks to the broad range of variation of the initial population (18ºBrix). Nevertheless, a reduction of the classification performance of the model has been observed with the passing of seasons. One of the reasons put forward is the reduction of the range of variation that naturally occurs during a breeding process, the other is the variations in other parameters than the variable of interest but whose effects would probably be affecting the measurements [1]. This study points to the application of Independent Component Analysis (ICA) on this highly variable dataset coming from a NIRS industrial application for the identification of the different sources of variation present through seasons.

More information

Item ID: 13405
DC Identifier: http://oa.upm.es/13405/
OAI Identifier: oai:oa.upm.es:13405
Official URL: http://www.chimiometrie.fr/chimiometrie2011/
Deposited by: Memoria Investigacion
Deposited on: 24 Oct 2012 08:26
Last Modified: 21 Apr 2016 12:43
  • Logo InvestigaM (UPM)
  • Logo GEOUP4
  • Logo Open Access
  • Open Access
  • Logo Sherpa/Romeo
    Check whether the anglo-saxon journal in which you have published an article allows you to also publish it under open access.
  • Logo Dulcinea
    Check whether the spanish journal in which you have published an article allows you to also publish it under open access.
  • Logo de Recolecta
  • Logo del Observatorio I+D+i UPM
  • Logo de OpenCourseWare UPM