Spoiled Food Detection Using a Matrix of Gas Sensors

The BME688 gas sensor interacts with the environment through the adsorption phenomenon. The metal-oxid’s membrane is sensitive to the surroundings based on the heating temperature of the membrane. The sensor returns ten resistance values according to the gas exposure. In the spoiled food detection, fruits and vegetables are used in mapping the digital olfactive footprint. Multiple recordings of fresh, almost fresh, almost spoiled, and spoiled states are acquisitioned in order to have sufficient data to be used in the neural algorithm. The resistance values are used as an input in a machine learning algorithm and the neural network labels if a food is fresh or spoiled.