Combining principal component analysis with an artificial neural network to perform online quality assessment of food as it cooks in a large-scale industrial oven

Marion O'Farrell, Elfed Lewis, Colin Flanagan, William B. Lyons, N. Jackman

Research output: Contribution to journalArticlepeer-review

Abstract

A sensor system utilising optical fibre sensing techniques is reported, which has been applied to the food industry in order to control the cooking process in a large-scale industrial oven. By monitoring a wide variety of products as they are cooked in the oven it has been possible to classify their cooking stage. This paper examines the application of principal component analysis, using karhunen loeve decomposition, to the spectral data from the sensor prior to application of pattern recognition through the use of an artificial neural network. Investigations have been carried out to ascertain trends in various products in order to design a general colour scale, which can classify several products using a single neural network. The food types discussed in this paper are steamed skinless chicken fillets, roast whole chickens, marinated chicken wings, sausages, pastry, breadcrumb coating and char-grilled chicken fillets.

Original languageEnglish
Pages (from-to)104-112
Number of pages9
JournalSensors and Actuators, B: Chemical
Volume107
Issue number1 SPEC. ISS.
DOIs
Publication statusPublished - 27 May 2005

Keywords

  • Artificial neural networks
  • Backpropagation
  • Cooked food
  • Food industry
  • Pattern recognition
  • Principal component analysis

Fingerprint

Dive into the research topics of 'Combining principal component analysis with an artificial neural network to perform online quality assessment of food as it cooks in a large-scale industrial oven'. Together they form a unique fingerprint.

Cite this