Dog Facial Recognition App Helps Scientists Fight Against Rabies

In countries where dogs aren’t microchipped, don’t wear tags, and usually run free, it’s challenging for authorities to determine which dog has already been vaccinated for rabies. A new mobile phone app could help improve rabies vaccination by identifying dogs using facial recognition technology.

A team of researchers from Washington State University tested the app’s effectiveness in a rabies vaccination clinic in Tanzania, where they microchipped, vaccinated, and registered dogs. 

The idea behind this project is to take a picture of a dog’s face during initial vaccination. Then the picture needs to be entered into the database via the app, along with other important information, such as the dog’s age, sex, and coat color. 

When veterinarians return to the same area, they will take a facial photo of each dog they come across. If the app finds a close match in the database, it will notify the vet, who can then decide whether the vaccinated dog in the database is indeed the same dog they are currently examining. 

A vaccination team testing the app visited nine villages in Tanzania, where they photographed and vaccinated 1,420 dogs. After the initial check, 20 images were deemed unusable, leaving exactly 1,400 to work with. All of these photographs were added to the app’s database.

Later, a different validation team visited the same villages, assessing 720 dogs via the app, not knowing which dogs were already vaccinated, and entered into the app’s database. Once again, some photos were deemed unusable for the study - 161 were rejected.

Using those images, the app helped the team to correctly identify 76.2% of vaccinated dogs and 98.9% of unvaccinated dogs. These findings were confirmed using microchips that were implanted under the dogs’ skin when they were vaccinated, as part of the study.

“Because domestic dogs are the main reservoir for human rabies, controlling human rabies globally requires the mass vaccination of dogs,” said Felix Lankester, WSU Associate Professor and the lead investigator of the study. 

Scientists are working to optimize the app’s registration process to minimize the number of rejected photos. 

“When carrying out mass vaccination, one of the major problems that we face is trying to identify which dogs have and haven’t been vaccinated. For example, microchips are too expensive to use at the scale needed to eliminate rabies, and collars can be removed by owners. We developed this app to see if facial recognition might work, and it’s showing great promise in helping us achieve that goal,” Professor Lankester added. 

The facial recognition algorithm used by the app is developed in collaboration with Canadian company PiP My Pet. It identifies a dog by looking at key features of its face and comparing it to previously stored photos of the faces of other dogs. Photos with the highest number of similar features are shown as possible matches, leaving it to the user to pick the right one.

The app’s success depends on the image quality and information about each dog being properly recorded. Before unusable images and incorrect information about dogs were removed from the database, users were only able to match around 65% of the vaccinated dogs.

For the time being, users must be online to use the app’s facial matching feature, but Lankester and his team are working to make the app operational even offline. 

“We’re not quite there yet, but I think with the investment, the technology can get there. I am excited by its potential,” Lankester said.

Join the PetGuide community. Get the latest pet news and product recommendations by subscribing to our newsletter here.