A team of researchers presents an artificial intelligence that differentiates between birds of the same species.

0
A team of researchers presents an artificial intelligence that differentiates between birds of the same species.
Photographie d’une Mésange charbonnière illustrant comment l’identification individuelle est faite par les ordinateurs. © André Ferreira

A team of scientists from CNRS/ University of Montpellier/ University Paul-Valéry-Montpellier/ IRD/ EPHE/ CIBIO at the University of Porto presented the results of their research on how computers can learn to differentiate individuals of the same bird species. Published on 27 July in Methods in Ecology and Evolution, the researchers used artificial intelligence techniques.

The recognition of different individuals of the same species is essential to study wildlife populations, their adaptation process and their behaviour. For the first time ever, research teams led by scientists from the Centre for Functional and Evolutionary Ecology (CNRS/ University of Montpellier/ Université Paul-Valéry-Montpellier/ IRD/ EPHE) and the CIBIO at the University of Porto identified individual birds using artificial intelligence techniques.

The researchers developed a technique that allowed them to collect a large number of photographs of birds: identified individuals bearing electronic tags were photographed from several angles.

This collection fed the computers, which then learned to recognise the birds by analysing the images using deep learning methods. Such a device allowed computers to distinguish individuals by the patterns on their plumage, a task that humans cannot perform. The machines were able to differentiate individuals in three species of birds, the Social Republican, the Sable Tit and the Mandarin Diamond.

This new process may not only lead to less invasive methods of identification of wild individuals, but also to new knowledge in ecology. AI could also open up new possibilities in the study of animal behaviour in natural populations.

Bibliography
Deep learning-based methods for individual recognition in small birds, André C. Ferreira, Liliana R. Silva, Francesco Renna, Hanja B. Brandl, Julien P. Renoult, Damien R. Farinec, Rita Covas and Claire Doutrelant, Methods in Ecology and Evolution on 27 July 2020. DOI: 10.1111/2041-210X.13436

Translated from Une équipe de chercheurs présente une intelligence artificielle différenciant les oiseaux d’une même espèce