Facial recognition can be used to identify primates and not humans alone, according to a new study published on May 29, 2018.
Researchers from Michigan State University adapted facial recognition technology to identify primates, and were successful in doing so with 93.76% level of accuracy. Tracking animals is beneficial for conservationists, as they are trying to protect animals. This recognition ensure an easy method of such conservation.
Several ways are currently being used to track animals, however, they are usually invasive. Facial recognition proves to be a non-invasive method and cost-effective as well. Traditional tracking devices cost between US$ 400 and US$ 4,000, while the actual tagging process is time-consuming and uncomfortable for animals involved.
PrimNet, a trained software by researchers is fed with thousands of reference photographs of primates. “Face detection also comes with some additional challenges due to the presence of variations in hair and fur, low contrast between eyes and background, and variation in eye colors across different individuals,” is explained in the paper.
This method improve the standard off the shelf human facial recognition software. PrimNet showed 93.76% accuracy for lemurs and 90.36% for golden monkeys, while only a maximum of 75.82% was recorded for Chimpanzees.
The accompanying Android app – PrimID – aims to resolve the accuracy crises altogether. If a photograph of a lemur is shot and it does not show 100% accuracy, then it offers information of five other likely candidates.