Automated Chilli Pesticide Residues Detection Using Odour Gas Sensors (OGS) and Deep Learning (DL) Algorithm

Detection of excessive pesticide residue detection is a serious problem for food regulators, suppliers, and consumers. It is very important to determine which chilli are contaminated with pesticides, and the current method of identifying and determining pesticide residues in chilli is still done using laboratory equipment. To overcome this problem, this study attempted to develop a method to detect pesticide residues in chilli samples using an eight different type of electronic nose based on a readily available metal oxide gas sensor. The proposed system used noise filtering, Long Short-Term Memory (LSTM) and Principal Component Analysis (PCA) algorithm along with a realtime data acquisition system that uses a computer to perform pesticide residue detection on the chilli sample. Two hundred forty samples of chilli sample with different pesticide concentrations went through the system and the accuracy rate achieved a success rate of 89.58% using the LSTM algorithm. The proposed method is expected to help the food processing industry to determine food contamination for producing clean and healthy food. The validation and feasibility of the proposed method for the determination of pesticide residues in chilli have been demonstrated by experiments.