Detection and analysis of the stored grain insect creeping sound

With the random acoustic source model as the theory model of the stored grain insects creeping, the sounds, of 20 Alphitobius diaperinus Panzer in wheat and 20 Tribolium castaneum Herbst adults in corn, are detected, respectively. By using Matlab, the original sound signals are reproduced and the de-noised signals are obtained. The power spectrums characteristics analysis are made. It is shown that the random acoustic source model is effective for the stored grain insect creeping sound detection, and their power spectrums are all discrete, where the highest frequency is 1600 Hz, the main frequency 205 Hz in the former, and the highest frequency is 800 Hz, the main frequency 350 Hz the latter, which may be used to distinguish different types insects in grain.