County shelter uses facial recognition to ID lost dogs
Miami-Dade County Animal Services has partnered with Finding Rover, a free mobile app that helps pet owners find their lost dogs using facial recognition technology.
Finding Rover maintains a database of dog facial images from registered app users, so when a lost dog’s picture is uploaded, the software compares that image with those in the database.
The app uses eight facial markers, far fewer than the 128 points on the human facial recognition program, to make the identification. It then locates the user via the phone’s GPS and starts searching for dogs that match the image of the lost dog in the area, said John Polimeno, the founder of Finding Rover.
Because Miami-Dade County’s Animal Services database of dogs is now integrated into the Finding Rover mobile app, users who have previously registered their pets can then search for found dogs at the shelter.
Polimeno said he worked with University of Utah for a year to develop the app, which uses facial recognition technology adapted from the fields of machine learning and computer vision. The Pet Match algorithm that researchers developed for the program detects key discriminating features in a dog’s face, such as the shape and position of the eyes, distinctive stripes or spots and the color of the fur, according to a report from the university.
“Additional information provided for the dogs, such as size, gender, breed and location, can be used prior to the facial recognition to narrow the collection of lost dogs Pet Match will consider for ranking,” said Steve Callahan, a software design engineer at the university’s Software Development Center.
Polimeno said more than 300 shelters have expressed interest in using the app, which is also in use in shelters in Houston and San Diego.
Posted by GCN Staff on Mar 19, 2015 at 10:13 AM