Practical discrimination of good and bad cooked food using metal oxide semiconductor odour sensor

Abstract

An increasing concentration of ammonia in cooked food is in direct proportion to the extent of decay. This fact is used to design an electronic nose (e-nose) based on metal oxide semiconductor odour sensor circuit capable of discriminating good and bad cooked food. On the basis of the data produced by the e-nose circuit, a feedforward multilayer neural network is designed and trained to recognize varying concentrations of ammonia in the food. Test results of the prototype e-nose system show that it is capable of classifying cooked food as being good or bad with over 92% average success rate.

Keywords

• neural network
• ammonia
• odour sensor
• metal oxide semiconductor
• electronic nose