Paddy and maize moldy status characterization using electronic nose.

摘要

Paddy and maize moldy status rapid detection using electronic nose (E-nose) was investigated in this paper. An E-nose system with eight metal oxide semiconductor gas sensors was used to measure moldy status of paddy and maize samples at room temperature. Principal component analysis (PCA) and stochastic resonance (SR) are used for data processing. PCA results showed that E-nose could distinguish paddy and maize samples in different moldy status. Signal-to-noise ratio (SNR) spectrum calculated by SR discriminated paddy and maize samples. Moldy status predicting models were developed based on SNR maximums linear fitting regression. Validating experiments demonstrated that the developed models presented high detecting accuracy. The proposed method is promising in grain quality rapid analysis applications.