HumBugDB: A Large-scale Acoustic Mosquito Dataset

Abstract

This paper presents the first large-scale multi-species dataset of acoustic recordings of mosquitoes tracked continuously in free flight. We present 20 hours of audio recordings that we have expertly labelled and tagged precisely in time. Significantly, 18 hours of recordings contain annotations from 36 different species. Mosquitoes are well-known carriers of diseases such as malaria, dengue and yellow fever. Collecting this dataset is motivated by the need to assist applications which utilise mosquito acoustics to conduct surveys to help predict outbreaks and inform intervention policy. The task of detecting mosquitoes from the sound of their wingbeats is challenging due to the difficulty in collecting recordings from realistic scenarios. To address this, as part of the HumBug project, we conducted global experiments to record mosquitoes ranging from those bred in culture cages to mosquitoes captured in the wild. Consequently, the audio recordings vary in signal-to-noise ratio and contain a broad range of indoor and outdoor background environments from Tanzania, Thailand, Kenya, the USA and the UK. In this paper we describe in detail how we collected, labelled and curated the data. The data is provided from a PostgreSQL database, which contains important metadata such as the capture method, age, feeding status and gender of the mosquitoes. Additionally, we provide code to extract features and train Bayesian convolutional neural networks for two key tasks: the identification of mosquitoes from their corresponding background environments, and the classification of detected mosquitoes into species. Our extensive dataset is both challenging to machine learning researchers focusing on acoustic identification, and critical to entomologists, geo-spatial modellers and other domain experts to understand mosquito behaviour, model their distribution, and manage the threat they pose to humans. Preprint. This paper has been accepted at the 35th Conference on Neural Information Processing Systems (NeurIPS 2021) Track on Datasets and Benchmarks.

Accurate Acoustic Based Deep Learning Model for Mosquitoes Species Classification

“…This slight improvement in accuracy could be attributed to the unique features of our model, which includes a use of sinc layer on raw audio data that focuses on filter parameters for better convergence. Additionally, unlike the existing methods which used laboratory data in a controlled environment to train and test their model, our proposed model was tested on real-time In another existing state of art research 17 , the model was trained and tested on a dataset of eight mosquito species, while our proposed model is evaluated on a dataset of four species. Despite this difference, our proposed model outperformed the previous research in terms of accuracy.…”

“…The HumBug project 17 is a global partnership between University of Oxford and mosquito experts from around the world. They collected 20 hours of mosquito sounds over five years from different locations and devices.…”

“…Acquiring wing beat data of different mosquitos in different real world data environments is a challenging task. The HumBug project 17 addresses this issue. As many as 36 distinct species have been annotated in 18 hours of audio recordings with varying SNR and background environments of Thailand, Kenya, Tanzania, USA and UK.…”