Using a reflection-based optical fibre system and Neural Networks to evaluate the quality of food in a large-scale industrial oven

A sensor system utilising optical fibre sensor techniques is reported for the online examination of food colour within large-scale industrial ovens. The various food products that are processed within these ovens are interrogated by employing spectroscopic techniques, with the resulting spectral patterns being interrogated and classified with the aid of Artificial Neural Networks. A system based on pattern recognition has been developed which is capable of classifying the various colours that can occur for each product into those which are favourable and those that are not optimum. This information can be used by the producer to control the cooking process online and optimise food quality. Spectral results are presented for a number of different food products, such as Sausages, Whole Chickens, Fresh Beef Burgers, High Cereal Patti Burgers and Marinated Chicken Thighs and Wings. This range of products was selected by Food Design Application Ltd. and was considered an adequate representation of the most popular food products cooked in the oven by their leading customers. A Neural Network was developed for each product and successfully classified the products into the following stages; raw, light correct and dark.

[1]  A. Treves,et al.  A neural network facial expression recognition system using unsupervised local processing , 2001, ISPA 2001. Proceedings of the 2nd International Symposium on Image and Signal Processing and Analysis. In conjunction with 23rd International Conference on Information Technology Interfaces (IEEE Cat..

[2]  E. Lewis,et al.  An intelligent optical fibre based sensor system for monitoring food quality , 2002, Proceedings of IEEE Sensors.

[3]  Elfed Lewis,et al.  Intelligent Processing of Spectroscopic Signals Obtained Using an Optical Fibre Based System for Food Quality Control , 2003 .

[4]  Suranjan Panigrahi,et al.  Computer-based neuro-vision system for color classification of french fries , 1995, Other Conferences.

[5]  Ronald Marsh,et al.  Conjugate gradient and approximate Newton methods for an optimal probabilistic neural network for food color classification , 1998 .

[6]  José Blasco,et al.  Assessment of citrus fruit quality using a real-time machine vision system , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[7]  Leonard G. C. Hamey,et al.  Colour bake inspection system using hybrid artificial neural networks , 1995, Proceedings of ICNN'95 - International Conference on Neural Networks.