Online pattern recognition, using ANN and SOM, to determine quality during the cooking process in the food industry

This paper reports on two methods of classifying the spectral data from an optical fibre based sensor system as used in the food industry. The first method uses a feed-forward back-propagation Artificial Neural Network while the second method involves using Kohonen Self-Organising Maps. The sensor monitors the food colour online as the food cooks by examining the reflected light, in the visible region, from both the surface and the core of the product The combination of using Principal Component Analysis (PCA) and backpropagation neural networks has been successfully investigated previously. In this paper, results obtained using this method are compared with results obtained using a Self-Organising Map trained on the Principal Components. PCA is performed on the reflected spectra, which form a "colourscale" - 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 colourscale data that can classify several food products. The results presented show that both classifiers perform well.