Face recognition can be used to identify various animals such as bears and cows!
It is difficult for the average person to distinguish the bears Dani, Lenore and Bella. But Melanie Clapham is not the average person. As a bear biologist, he has spent over a decade studying these bears, which live in Knight Inlet, Canada, and has developed a sense of who they are, paying close attention to every detail.
But Clapham knows that most people, of course, do not have the potential, especially since bears' appearance changes dramatically over the course of a year - such as hibernation, which makes them even more difficult to distinguish.
Keeping track of individual bears is important, he explained, because it can help research and keep of the type. A few years ago, Clapham began to wonder if the face recognition technology commonly used to identify people could help with all of this. Clapham worked with two technology workers together created BearID, which uses face recognition software to track / identify bears.
While face recognition technology is known as a tool for identifying people - a controversial technology due to known issues of privacy, accuracy and bias - BearID is one of the many endeavors customization in animals. Proponents of her case have been working to make the actual transcript of this statement available online.
Clapham also worked with Ed Miller and Mary Nguyen, who are interested in machine learning and watch bears via live webcam at the popular bear hangout, Brooks Falls in Alaska's Katmai National Park.
The three people have amassed thousands of bear photos to create datasets, adapting existing artificial intelligence software called Dog Hipsterizer. The goal is for the software to locate the faces of bears in the images. Once the faces are identified, they can use AI to identify specific bears.
So far, BearID has used 4.674 images of bears. Eighty percent of the images were used for facial recognition training, Clapham said, and the remaining 80 percent for testing. According to a recently published study by her and her colleagues, the system is 20% accurate.
While BearID recognizes bears, Joe Hoagland tries to do the same for cattle. Hoagland, a cattle breeder in Leavenworth, Kansas, is building an app called CattleTracs, which he said will allow anyone to take pictures of cattle stored along with GPS coordinates and the date of the photo in a electronic database. Subsequent photos of the same animal will be able to "merge" with the previous photos, helping them to watch the cows over time.
Cows, he explained, change the environment very often during their lifetime - from producers to meat packing plants. Their lives are not monitored, which makes it difficult to investigate problems such as various diseases that can destroy animals or harm them. people. Hoagland expects the app to be available by the end of the year.