Facial Recognition for Wild Primate Faces
Facial recognition software, normally developed to recognise the faces of human individuals, to interact with and electronically monitor the activities of individuals, has been reapplied to study other animals, keeping the same basic technology in-place.
Why not? After all, other animals have individual differences in their faces as well identifying marks, different widths of nose, length of brow ridge, shape of the eyes, mouth or chin. The software should be a good fit, if there's ever a need to track an individual animal, monitor their activities, or interact with them specifically.
One immediate need that crops up, is the protection of endangered species. In this case, great apes. Game wardens studying the behaviour of surviving great apes in the wild, have a difficult time of it. This is painstaking, often backbreaking work, tracking individuals, and distinguishing between them short of tracking every single individual with electronic tags in real-time; something that is all but impossible due to current practical constraints.
So, why not use facial recognition technology and video traps instead? Work out from the faces of the great apes who appear in the area, which individuals are present, along with noticing any new ones, or the change when a given individual no-longer turns up? This is not tagging the animals, with an electronic signal a poacher could home in on, but instead tagging the areas themselves and determining who passes through. If a poacher homes in on that signal, they'll walk straight into the hands of the authorities as well as get their face on tape.
The system, developed as part of the SAISBECO project, was developed by a partnership between the Fraunhofer Institutes for Integrated Circuits IIS and Digital Media Technology IDMT as well as at the Max Planck Institute for Evolutionary Anthropology.
"The pictures are first filtered to find the ones on which the primates' faces can be seen," explains Alexander Loos from the IDMT in Ilmenau. This task is handled by a detection software program developed by research scientists at the IIS in Erlangen which detects faces on individual pictures as well as on video streams in real time. Loos and his colleagues are currently developing a module that will assign the faces to specific individuals. "Our software analyses the primates' faces using special algorithms," states Loos. At the moment these face-recognition algorithms analyze the entire face. In a data pool of 24 chimpanzees at Leipzig Zoo the Max Planck research scientists achieved a recognition rate of 83 percent.
This good hit accuracy is due to the high quality of the photos. "The algorithms are strongly affected by external influences," explains Loos. "In poor light or if the faces are partially occluded the recognition rates quickly drop to below 60 percent." Because it is much more difficult to get good pictures in the wild, the Ilmenau-based research scientists intend to add further algorithms which will not analyze the entire face but specific biometric features -- such as the eyes, nose and mouth.
A second software process analyses analyses audio signals and assigns them to various noises made by the apes, for example chest drumming and threatening grunts like pant-hoots. In other words, it is augmenting the detection process by means of behaviometrics working out who the individual is by specific behavioural traits and idiosyncrasies.