An Electronic Tool to Differentiate Between Potatoes According to Fertilization Methods

The chapter concentrates on the distinction between two varieties of potatoes: the first kind is harvested from soil traditionally processed with manure from domestic sheep and donkeys, and the second kind is harvested from soil processed with chicken farms’ manure utilizing a multi-sensor. The hardware part of our electronic nose (multi-sensor) tool is made principally of five commercial gas sensors, where the sensing element is a metal oxide. For the data analysis part and after preprocessing, the treatment of the data results from the sensors was done first using the principal component analysis (PCA) pattern, the first three principal components were able to explain 99.20% of the variance. Subsequently, the K-nearest neighbors (KNNs) method was used for classification with fivefold cross-validation as the type of evaluation model. A high success rate of 95% was attained by this method which will allow us to differentiate properly between the two kinds of potatoes.