Mechanical operations, like sowing and plowing, inadvertently destroy a huge number of ground-nests of farmland birds every spring.
Researchers from the University of Helsinki have performed a new study that shows, for the first time, that such nests can be identified with the help of a combination of a drone and artificial intelligence.
In a majority of the regions in Europe, farmland bird species are declining. Birds that breed on the ground are specifically susceptible since they are prone to mechanical operations, such as sowing and plowing, which occur during spring and accidentally ruin the nests.
Locating Nests on the Ground is challenging for the Human Eye, and Highly Time-Consuming
A drone with a thermal camera was flown by the team over the agricultural fields to capture images. Then, these images were fed to an artificial intelligence algorithm with the ability to find out the nests precisely, an initial step in ensuring their safety. The team tested the system in Southern Finland next to the University of Helsinki’s Lammi Biological Station, by making use of the wild nests with eggs of the Lapwing Vanellus vanellus.
We have been involved in conservation of ground-nesting farmland birds for years and realized how difficult it is to locate nests on the ground. At least at high latitudes, the temperature of these nests is typically higher than that of the surrounding environment. Hence, we thought that thermal cameras could assist.
Andrea Santangeli, Academy of Finland Fellow, Finnish Museum of Natural History Luomus, University of Helsinki
Santangeli continued, “A small pilot study indicated that thermal vision is hampered by vegetation and objects on the ground. Therefore, to make this an efficient system, we thought that the camera could be flown using a drone, and artificial intelligence could help to analyze the resulting thermal images. We show that this works. However, the system performed best under cloudy and cold conditions, and on even grounds.”
Drone Technology Becoming Rapidly Popular in Conservation
With the help of drones powered by several sensors, the spread of diseases on crops in agricultural regions can be mapped in near real-time. Drones form an integral part of precision agriculture, a new crop production technique that involves the widespread use of drone technology to supervise the crops and optimize production efficiency.
Such studies can help provide a method to incorporate bird nest detection inside the drone-borne sensors that have been utilized in precision agriculture and automate a system for rescuing those nests.
The conservation community must be ready to embrace technology and work across disciplines and sectors in order to seek efficient solutions. This is already happening, with drone technology becoming rapidly popular in conservation.
Andrea Santangeli, Academy of Finland fellow, Finnish Museum of Natural History Luomus, University of Helsinki
Santangeli added, “A next and most challenging step will be to test our system in different environments and with different species. Our auspice is that this system will be, one day, fully integrated into agricultural practices, so that detecting and saving nests from mechanical destruction will become a fully automated part of food production.”
Santangeli, A., et al. (2020) Integrating drone-borne thermal imaging with artificial intelligence to locate bird nests on agricultural land. Scientific Reports. doi.org/10.1038/s41598-020-67898-3.