ENOSE FOR INTERNET OF THINGS

Abstract

System “eNose” is developed within a project “eNose for Internet of things”, which is a part of a project contest “Research Grant of Rezekne Acadaemy of Technologies”. The aim of this work is to explore whether it is possible to detect spoiled food with help of sensors and a neural network. System “eNose” is intended to detect and classify spoiled food products within storages and notify related users. Detection and classification are performed by four gas sensors and a neural network. As a result, a web application was developed that performs such functions as storage and sensor registration, neural network training, spoiled food detection based on sensor data, and user permission control. It was concluded that sensors for this application must be very precise in order to receive best possible results.