AI to the rescue: technology can protect nesting birds

Birds nesting on the ground across farmlands in Europe often face a singular threat: plows and other agricultural tools. Each spring numerous breeding farmland birds fall victim to agricultural activities as people fail to spot them in time before destroying their nests by accident.

Yet science can come to the rescue in the form of drones and artificial intelligence.

A team of researchers from the University of Helsinki decided to fly a drone equipped with a thermal camera over some agricultural fields in southern Finland, then fed the resulting images to an AI algorithm designed to identify nests of northern lapwings (Vanellus vanellus).

During a pilot study the researchers found that thermal vision when use at ground level was hampered by the presence of dense vegetation and objects in the way. So they decided to give the camera a bird’s eye view by making it airborne with a drone.

The technique worked like a charm. The thermal imaging system works best on cloudy days and when the temperature is colder. “At least at high latitudes, the temperature of these nests is typically higher than that of the surrounding environment,” explains Andrea Santangeli, an fellow at the Finnish Museum of Natural History Luomus at University of Helsinki.

The technology is far superior to human eyes, which can be a big bonus in protecting threatened birds that are fast losing their habitats to agricultural activities, Santangeli says. “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,” he notes.

Drones equipped with sensors are already in use in precision agriculture for mapping the spread of diseases on crops and monitoring other threats to them. The new AI technology could now be employed effectively in conservation efforts such as “by integrating nest detection within the precision agriculture system that heavily relies on drone-borne sensors,” the scientists explain in a study on their findings.

“The conservation community must be ready to embrace technology and work across disciplines and sectors in order to seek efficient solutions,” Santangeli stresses. ” This is already happening, with drone technology becoming rapidly popular in conservation.”

The next step involves fine-tuning the system for use in other environments to protect other threatened species. The scientists hope that soon their system will be “fully integrated into agricultural practices, so that detecting and saving nests from mechanical destruction will become a fully automated part of food production.”