Have you ever looked at a seal and thought, Is it the same seal I saw yesterday? Well, there could soon be an app for this based on the new seal face recognition technology. Known as SealNet, this seal face tracking system was developed by a group of undergraduate students from Colgate University in New York. Inspired by other technology adapted for the identification of primates and bears, Krista Ingram, a biologist at Colgate University, led the students to develop software that uses deep learning and a cohesive neural network to distinguish seals from one another. SealNet is customized to identify port seals, a species that tends to pose on shores on excursions. The team had to train its software to identify seal faces. “I give him a picture, he finds the face, [and] it fits into a standard size, “says Ingram. But then she and her students would manually locate the nose, mouth and center of the eyes. For the project, team members took more than 2,000 photos of seals around Casco Bay, Maine, over a two-year period. They tested the software using 406 different seals and found that SealNet was able to correctly identify seal faces in 85 percent of cases. Since then, the team has expanded its database to include about 1,500 seal individuals. As the number of stamps registered in the database increases, so will the accuracy of identification, says Ingram. SealNet developers trained a neural network to distinguish port seals using photos of 406 different seals. Photo courtesy of Birenbaum et al. As with all technology, however, SealNet is not infallible. The software saw seal faces on other parts of the body, vegetation and even rocks. In one case, Ingram and her students made a double impression of the incredible resemblance between a rock and a seal. “[The rock] “He looked like a seal,” says Ingram. “The darkest parts were about the same distance as the eyes; so you can understand why the software found a face.” Therefore, it says it is always best to manually check that the stamp faces identified by the software belong to a real stamp. Like a tired seal crawling on a beach for an inadvertent photo shoot, the question arises as to why all this is necessary. Ingram believes that SealNet could be a useful, non-invasive tool for researchers. Of the world’s finches – a group that includes seals, seahorses and sea lions – harbor seals are considered to be the most widespread. However, there are knowledge gaps. Other seal tracking techniques, such as tagging and aerial tracking, have their limitations and can be extremely invasive or costly. Ingram cites the site’s loyalty as an aspect of seal behavior that SealNet could shed more light on. The team’s tests showed that some port seals return to the same transport sites from year to year. Other seals, however, such as two animals nicknamed Clove and Petal by the group, appeared in two different locations together. Scientists’ growing understanding of how seals move could strengthen arguments for protecting specific areas, says Anders Galatius, an environmentalist at the University of Aarhus in Denmark who did not participate in the project. Galatius, who is in charge of monitoring Danish seal populations, says the software “shows a lot of promise”. If the recognition rates improve, it could be combined with another photo recognition method that identifies seals with distinctive marks on their ax, he says. In the future, after further testing, Ingram hopes to develop an application based on SealNet. The application, he says, could potentially allow citizen scientists to help record seal faces. The program could also be adapted for other winged animals and possibly even cetaceans.
