Peer-Reviewed Journal Details
Mandatory Fields
Lewis, E,Sheridan, C,O'Farrell, M,Flanagan, C,Kerry, JF,Jackman, N
Nonlinear Analysis-Hybrid Systems
Optical fibre sensors for assessing food quality in full scale production ovens - a principal component analysis and artificial neural network based approach
Optional Fields
Optical fibre temperature sensor Food quality sensor Optical fibre colour sensor Artificial neural networks processing for sensor data PCA
This paper reports on a method of classifying the spectral data from an optical fibre based sensor system as used in the food processing industry for monitoring food products as they are cooked in large scale continuous ovens. The method uses a feed-forward back-propagation artificial neural network. The sensor monitors the food colour online as the product cooks by examining the reflected light, in the visible region, from both the surface and the core of the product. Results based on the combined use of Principal Component Analysis (PCA) and standard back-propagation artificial neural networks are presented. Results are also reported for a wide range of food products which have been cooked in the full scale industrial oven. PCA is performed on the reflected spectra, which form a "colour scale" - a scale developed to allow the quality of several products of similar colour to be monitored, i.e. a single classifier is trained, using the colour scale data, that can classify several food products. The results presented show that the classifier performs well. (c) 2007 Elsevier Ltd. All rights reserved.
DOI 10.1016/j.nahs.2006.05.008
Grant Details