Resolving the identification of weak‐flying insects during flight: a coupling between rigorous data processing and biology. (2nd June 2021)

• Record Type: Journal Article
• Authors: Hassall, Kirsty L.; Dye, Alex; Potamitis, Ilyas; Bell, James R.
• Is Part Of: Agricultural and forest entomology. Volume 23:Number 4(2021)
• Page Start: 489
• Page End: 505
• Publication Date: 2021-06-02
• Subjects: Aphid -- beetle -- insect classification -- random forest classification -- wingbeat harmonics; Insect pests -- Control -- Periodicals; Agricultural pests -- Control -- Periodicals; Forest insects -- Control -- Periodicals; 632.7
• DOI: 10.1111/afe.12453
• Languages: English
• ISSNs: 1461-9555

Abstract

Abstract: Bioacoustic methods play an increasingly important role for the detection of insects in a range of surveillance and monitoring programmes. Weak‐flying insects evade detection because they do not yield sufficient audio information to capture wingbeat and harmonic frequencies. These inaudible insects often pose a significant threat to food security as pests of key agricultural crops worldwide. Automatic detection of such insects is crucial to the future of crop protection by providing critical information to assess the risk to a crop and the need for preventative measures. We describe an experimental set‐up designed to derive audio recordings from a range of weak‐flying aphids and beetles using an LED array. A rigorous data processing pipeline was developed to extract meaningful features, linked to morphological characteristics, from the audio and harmonic series for six aphid and two beetle species. An ensemble of over 50 bioacoustic parameters was used to achieve species discrimination with a success rate of 80%. The inclusion of the dominant and fundamental frequencies improved prediction between beetles and aphids because of large differences in wingbeat frequencies. At the species level, error rates were minimized when harmonic features were supplemented by features indicative of differences in species' flight energies. Abstract : Data processing pipeline for the automatic identifcation of weak‐flying insects. Insects fly through an opto‐acoustic sensor triggering an audio recording to be made of the sound of their flight. Data are then processed to enable feature extraction. These features are fed into classification algorithms, namely random forests, which are then checked for validity before being sent to the end user for interpretation.