Spectral Model for Activity Monitoring of Red Palm Weevil

摘要

Pests’ detection and monitoring in fields and orchards is essential for effectivecontrol. Common monitoring methods are visual, and not effective in detectinghidden pests. The red palm weevil (Rhynchophorus ferrugineus, Olivier), a native pestfor various South-East Asia palm species, is presently considered the most importantpest of date palms in the Middle East. The weevil larvae, while boring deep in palmtrunk, are causing lethal damage to the tree by destroying its vascular bundles. Datepalms are commercially propagated by offshoots, the hidden weevil larvae in theoffshoots are transported this way between and within countries. Our experimentsindicated the possibility of acoustically monitoring the red palm weevil activity indetached palm offshoots using a dedicated microphone, signal amplification andprocessing.Acoustic activity was recorded from offshoots, potted palms infested with the weevillarvae and from uninfected ones. Sample sound clips, of equal length, were extractedfrom the recordings using an interactive tool that was developed under Matlabenvironment. The clips were manually separated to a set that included apparent larvaactivity, and another set of clips with no pest activity.Every clip was analyzed using FFT to extract the amplitude in the frequency domain.Apparent distinction appeared between the line spectra of clips with and without