A new app utilizing facial recognition technology allows for the identification of dogs using a mobile device. It will be used as part of rabies control programs in countries in Africa and Asia.
Researchers from Washington State University (WSU) tested the app at a rabies vaccination clinic in Tanzania, where they conducted microchipping, vaccination, and registration of dogs. According to them, the app was able to identify 71.2% of vaccinated animals and 98.9% of unvaccinated ones.
“One of the main challenges we face in mass vaccination is determining which dogs have been vaccinated and which have not. For example, microchips are too expensive to use on the scale needed to eradicate rabies, and collars can be removed by owners. We developed this app to see if facial recognition technology could help solve this problem, and it showed great potential,” said the study leader, WSU Assistant Professor Felix Lankester. His comments were published on the university's official website.
It is reported that the facial recognition algorithm used in the app was developed by WSU in collaboration with the Canadian company PiP My Pet. It identifies a dog by capturing key points of its face and matching them with images of other dogs' faces stored in a corresponding database. It is noted that the accuracy of the app's performance depends on the quality of the image and the correctness of the data about each dog, such as age, color, and sex.
Currently, the app requires internet access for proper functioning. However, according to Lankester, there are plans to implement an offline mode, as network access is not always available in remote areas.
Scientists emphasize that around 60,000 people die from rabies each year. Most cases occur in Africa and Asia, and more than 99% of them result from dog bites.
The results of the experiment using the app were published in the journal Scientific Reports.
