New research demonstrates for the first time that artificial intelligence (AI) can be used to train computers to recognise individual birds, a task humans are unable to do. The research is published in the British Ecological Society journal Methods in Ecology and Evolution.
“We show that computers can consistently recognise dozens of individual birds, even though we cannot ourselves tell these individuals apart. In doing so, our study provides the means of overcoming one of the greatest limitations in the study of wild birds - reliably recognising individuals without external tags.” Said André Ferreira, a PhD student at the Centre for Functional and Evolutionary Ecology (CEFE), France, and lead author of the study.
In the study, researchers from the FitzPatrick Institute of African Ornithology based at UCT, the French National Centre for Scientific Research (CNRS), and other institutes from Portugal (CIBIO) and Germany (Max Planck) describe the process of collecting thousands of labelled images of birds and then using this data to train and test AI models. This study represents the first successful attempt to do this with birds.
The researchers trained the AI models to recognise images of individual birds in wild populations of sociable weavers, great tits and a captive population of zebra finches. After training, the AI models were tested with images of the individuals they had not seen before and had an accuracy of over 90% for the wild species and 87% for the captive zebra finches.
In animal behaviour studies, individually identifying animals is one of the most expensive and time-consuming factors, limiting the scope of behaviours and the size of the populations that researchers can study. Current identification methods like attaching colour bands to birds’ legs can also be stressful to the animals.
These issues could be solved with AI models. Claire Doutrelant, CNRS researcher and director of this study, said: “The development of methods for automatic, non-invasive identification of animals completely unmarked and unmanipulated by researchers represents a major breakthrough in Conservation, Ecology and Evolution studies. It will open plenty of room to find new applications for this system and answer questions that seemed unreachable in the past. Rita Covas (UCT-CIBIO) and I will use it on sociable weavers, one of the most cooperative species of the world, to determine which bird invest in the common good (the gigantic nest mass) and to determine how kin, social and sexual selection could lead to the evolution of this cooperative behaviour.”
For AI models to be able to accurately identify individuals, they need to be trained with thousands of labelled images. Today human recognition is possible because models have access to millions of pictures of different people that are voluntarily tagged by users. But, acquiring such labelled photographs of animals was not possible before and has created a bottleneck in research.
André Ferreira was able to overcome this challenge by building feeders with camera traps and sensors. Most birds in the study populations carried a passive integrated transponder (PIT) tag, similar to the microchips implanted in pet cats and dogs. Antennae on the bird feeders were able to read the identity of the bird from these tags and trigger the cameras that took hundreds of photos.
Being able to distinguish individual animals from each other is essential for the long-term monitoring of populations and protecting species from pressures such as climate change. It is also essential to have access to behaviours that were never described before and understand their evolution and consequences.
André C. Ferreira et al, Deep learning-based methods for individual recognition in small birds, Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society, 2020