An acoustic system for the individual recognition of insects

Abstract

Research into acoustic recognition systems for insects has focused on species identification rather than individual identification. In this paper, the feasibility of applying pattern recognition techniques to construct an acoustic system capable of automatic individual recognition for insects is investigated analytically and experimentally across two species of Orthoptera. Mel-frequency cepstral coefficients serve as the acoustic feature, and α-Gaussian mixture models were selected as the classification models. The performance of the proposed acoustic system is promising and displays high accuracy. The results suggest that the acoustic feature and classifier method developed here have potential for individual animal recognition and can be applied to other species of interest.

Automated classification of bees and hornet using acoustic analysis of their flight sounds

“…There were clear differences in the frequency spectra and the harmonic components between their flight sounds and the background sounds ( Figure 1). Therefore, we used mel-frequency cepstral coefficients (MFCC) to describe the acoustic characteristic feature values of the different types of sounds, because MFCC was one of the most frequently used feature values in identifying sounds from different insects in previous studies, such as Orthoptera (Chaves et al 2012;Zhang et al 2012), Cicadae (Zilli et al 2014), and some bumble bees (Gradišek et al 2017). Basically, MFCC describes the timbre of sounds and is calculated using the following steps: (1) slicing the original sound into frames, (2) applying a window function to each frame, (3) applying Fourier transformation to each frame and obtaining the power spectrum of each frame, (4) applying mel-scale filter banks to the frames, and (5) applying a discrete cosine transformation (DCT).…”