Application of PCA Method on Pest Information Detection of Electronic Nose

In this paper, we apply electronic nose to detect crop pest information for the first time, based on the obtained sensor array data. Feature parameters from each sensor curve such as maximum, max differential value, mean value and stable value etc. are extracted and then used as the input of pattern recognition, then principal component analysis (PCA) is adopted to analyze the test sample. Experiments investigate the PCA method on electronic nose is able to detect whether rice is attacked by insect pests, to know the inroad extent of damaged rice and the amount of pests on each stem of paddy rice.