Mosquito Detection with Neural Networks: The Buzz of Deep Learning

TL;DR

This paper demonstrates that deep learning, specifically CNNs on wavelet-transformed audio, can effectively detect mosquitoes acoustically in scarce data scenarios, outperforming traditional methods and human experts.

Contribution

It introduces a CNN-based approach for acoustic mosquito detection using wavelet features, providing insights into feature importance and surpassing existing methods.

Findings

CNNs outperform conventional classifiers in mosquito detection

Deep features learned by CNNs are more informative than hand-crafted features

Detection accuracy exceeds existing algorithms and marginally surpasses human experts

Abstract

Many real-world time-series analysis problems are characterised by scarce data. Solutions typically rely on hand-crafted features extracted from the time or frequency domain allied with classification or regression engines which condition on this (often low-dimensional) feature vector. The huge advances enjoyed by many application domains in recent years have been fuelled by the use of deep learning architectures trained on large data sets. This paper presents an application of deep learning for acoustic event detection in a challenging, data-scarce, real-world problem. Our candidate challenge is to accurately detect the presence of a mosquito from its acoustic signature. We develop convolutional neural networks (CNNs) operating on wavelet transformations of audio recordings. Furthermore, we interrogate the network's predictive power by visualising statistics of network-excitatory samples. These visualisations offer a deep insight into the relative informativeness of components in the detection problem. We include comparisons with conventional classifiers, conditioned on both hand-tuned and generic features, to stress the strength of automatic deep feature learning. Detection is achieved with performance metrics significantly surpassing those of existing algorithmic methods, as well as marginally exceeding those attained by individual human experts.