Mothbox: inexpensive, lightweight, automated light trap for scalable insect biodiversity monitoring
Szczygiel, H.; Dent, D.; Quitmeyer, A.
Show abstract
O_LIInsects represent the most diverse group of organisms on Earth, and comprise the majority of known species, yet they are seldom accounted for in large-scale biodiversity monitoring systems and conservation planning. C_LIO_LIWe have developed the Mothbox: an automated light trap that uses open science hardware to make insect monitoring accessible to non-specialists and scalable for scientific and conservation purposes. We have also developed Mothbot: a machine learning tool and user interface for processing data from automated light traps. The Mothbox is portable, durable to harsh environments, and low cost to produce, while Mothbot prioritizes human-in-the-loop data validation of computer vision outputs to ensure data quality and aid in developing custom image reference libraries. C_LIO_LIMothbox has now been extensively tested in field conditions; with > 185 deployments, spanning > 450 nights and > 100 sampling locations. The two case studies presented here demonstrate the potential of Mothbox data for behavioural studies and conservation projects. C_LIO_LIAs the young field of automated insect monitoring continues to develop, Mothbox provides an affordable entry point for a wide range of stakeholders. It enables entomologists to explore new research questions by tracking insect diversity in real time and across entire landscapes. Mothbox will help build global insect-monitoring capacity, improving our ability to detect biodiversity loss and guide effective conservation action. C_LI
Matching journals
The top 2 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Improving the accessibility and transferability of machine learning algorithms for identification of animals in camera trap images: MLWIC2 94%
- Forecasting the Effects of Global Change on a Bee Biodiversity Hotspot 93%
- Monitoring free-living honeybee colonies in Germany: Insights into habitat preferences, survival rates, and Citizen Science reliability 92%
Similar papers in this journal
- A workflow for microclimate sensor networks: integrating geographic tools, statistics, and local knowledge 94%
- Using machine learning to count Antarctic shag (Leucocarbo bransfieldensis) nests on images captured by Remotely Piloted Aircraft Systems 93%
- Comparison of Two Individual Identification Algorithms for Snow Leopards (Panthera uncia) after Automated Detection 93%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.