Electronic Nose (E-Nose)

摘要

Smell is one of the six senses of human beings and there is not a lot of work on that sense, therefore, this paper is based on that sense. It aims to develop Electronic Nose (E-Nose) that will work not only to determine the ripeness level of fruits but also to make prediction that how long they will remain in that particular stage and also the classification of fruits as well, through smell. In order to achieve that K-Nearest Neighbors (KNN) algorithm is used. Hardware components include ME3-C2H4 sensors, Raspberry pi 3. The researchers work almost 10 hours daily to get the dataset of ethylene gas as accurate as possible. The dataset will then be used in KNN algorithm for classification (to determine the name) & categorization (to determine the ripeness level) of fruits. On the basis of experiments the success rate of this system is 70%.