Insect Detect: An open-source DIY camera trap for automated insect monitoring

Abstract

Insect monitoring is essential to design effective conservation strategies, which are indispensable to mitigate worldwide declines and biodiversity loss. For this purpose, traditional monitoring methods are widely established and can provide data with a high taxonomic resolution. However, processing of captured insect samples is often time-consuming and expensive, which limits the number of potential replicates. Automated monitoring methods can facilitate data collection at a higher spatiotemporal resolution with a comparatively lower effort and cost. Here, we present the Insect Detect DIY (do-it-yourself) camera trap for non-invasive automated monitoring of flower-visiting insects, which is based on low-cost off-the-shelf hardware components combined with open-source software. Custom trained deep learning models detect and track insects landing on an artificial flower platform in real time on-device and subsequently classify the cropped detections on a local computer. Field deployment of the solar-powered camera trap confirmed its resistance to high temperatures and humidity, which enables autonomous deployment during a whole season. On-device detection and tracking reliably estimated insect activity/abundance after metadata post-processing. Our insect classification model achieved a high top-1 accuracy on the test dataset and generalized well on a real-world dataset with captured insect images. The camera trap design and open-source software are highly customizable and can be adapted to different use cases. With custom trained detection and classification models, as well as accessible software programming, many possible applications surpassing our proposed deployment method can be realized.

Towards a toolkit for global insect biodiversity monitoring

“…Recently, several camera trap designs have been developed to monitor insects at low maintenance costs (e.g. [ 37 – 39 ]). In parallel, imaging devices for laboratory studies of insect specimens are being developed that can efficiently capture rich information from bulk samples and single specimens [ 30 , 40 , 41 ].…”

Utilising affordable smartphones and open-source time-lapse photography for monitoring pollinators

“…To mitigate the heat issue, we used white casings that reflect light (a simple solution we employed involved using a piece of paper). This issue may also arise in microcomputers, where heatsinks and coolers are essential to maintain the device at operational temperatures (e.g., Sittinger et al 2023). Nevertheless, surveillance cameras designed for outdoor use appear to be more resilient to such conditions, though they are typically more expensive than smartphones (pers.…”

“…Although we configured the Open Camera app to take an image every second, in practice, due to overheating or processing power limitations, smartphones struggle to achieve this rate, typically capturing one image every 1.5 to 2 seconds. In contrast, microcomputers can rapidly capture dozens of images per second (e.g., Droissart et al 2021; Sittinger et al 2023) without requiring significant processing power. Nonetheless, even at reduced frame rates of 1.5 or 2 seconds, we were typically able to capture multiple images of each visiting individual, enabling taxonomists to identify insects by observing them from various angles.…”

“…However, several challenges exist, such as insects not emitting heat, often being too small to trigger motion sensors, or sensors being activated by wind movement. There are ongoing efforts to develop AI triggers for pollinators, which involve real-time AI pipelines (edge AI) powered by nano GPUs in the field (e.g., Bjerge et al 2022; Sittinger et al 2023).…”

The FAIR-Device - a non-lethal and generalist semi-automatic Malaise trap for insect biodiversity monitoring: Proof of concept

“…Recently developed non-invasive systems that utilize cameras and AI algorithms for insect detection and identification provide an exception to lethal automated monitoring devices. For instance, the Insect Detect DIY camera trap is designed to video-monitor a colored surface through tracking and recognition algorithms [11, 12]. This surface is optimized to attract mainly hoverflies ( Syrphidae ) and Hymenoptera .…”