Electronic tongue and electronic nose data fusion in classification with neural networks and fuzzy logic based models

Abstract

One of the most interesting application areas of electronic noses is the food industry. In some cases when the results are not satisfactory, fusing the data of an electronic nose and an electronic tongue can result in highly increased performance. In this paper we combine the information of both instruments, and test their performance in potato chips and potato creams classification problem. Results for five classification techniques are compared, all based either on fuzzy logic systems or artificial neural networks. The results obtained using just nose or tongue information are compared to those when both instruments were fused. We show that the overall performance of a classifier was substantially increased for all algorithms.