Emerging technologies support a new era of applied wildlife research, generating data on scales from individuals to populations. Computer vision methods can process large datasets generated through image-based techniques by automating the detection and identification of species and individuals. With the exception of primates, however, there are no objective visual methods of individual identification for species that lack unique and consistent body markings.
With biodiversity conservation a global concern, computational approaches that enable wildlife monitoring at larger spatial scales, but with finer resolution, are recognized as a priority. Computer vision increasingly supports analyses of big data collected from image-based ecological studies. One challenge is the inability to distinguish among individuals within species that lack unique markings. Addressing this gap and taking a similar approach to human individual identification, face recognition has been developed for nonhuman primates. For species other than primates, one of the only references to facial recognition of unmarked species in the peer-reviewed literature focuses on domestic dogs Canis familiaris.
Facial recognition approaches could prove useful to the suite of non-primate wildlife species that lack distinctive body markings. Knowledge of unique individuals can facilitate the use of established techniques such as mark–recapture and thereby inform management.
Goose Facial Recognition
The recent emergence of advanced facial recognition techniques in the animal kingdom is revolutionizing the way researchers track and study various species. NPR’s Geoff Brumfiel reported on this fascinating development. He talked about how AI is now being applied to the faces of geese, harbor seals and other animals.
Sonia Kleindorfer, a biologist at the University of Vienna and the director of the Konrad Lorenz Research Center for Behavior and Cognition journeyed into the world of goose facial recognition, inspired by her role at the research center. Following the work of Austrian biologist Konrad Lorenz, who dedicated much of his career to studying the behavior of grayling geese, she quickly realized that distinguishing between individual geese based on their faces was no easy feat.
Kleindorfer created a database of geese photos from every conceivable angle, based on which they developed a facial recognition AI system capable of identifying geese by analyzing specific features of their beaks. Although the process took several years, the goose recognition software now boasts an impressive accuracy rate of approximately 97%, as reported in the Journal of Ornithology.
The facial recognition technology in the animal kingdom extends beyond geese. Researchers have applied this technology to a diverse range of species, from lemurs to bears, with promising results.
Future of Wildlife conservation
The researchers believe that facial recognition technology will play a pivotal role in the future of wildlife conservation and ecology. Researchers will be able to determine the size of populations, understand social interaction and monitor individual movements more efficiently. However, for this potential to be fully realized, collaboration between computer scientists and conservation biologists will be essential.
The integration of facial recognition technology into the study of animal behavior marks a significant milestone in wildlife conservation and ecology. With the ability to identify and track individual animals more accurately and efficiently, researchers are poised to make groundbreaking discoveries that will benefit not only our understanding of wildlife but also our efforts to protect and conserve these precious species.