Photo Recognition Software, New Tool for Understanding Harbor Seals

I don’t know about you but I find harbor seals irresistible. They are one reason I love to kayak in Casco Bay. I never know when one of their roundish heads will appear on the surface. We acknowledge each others presence then the seal will point it’s nose skyward (I like to think of it as a salute, then disappear below the surface. On occasion, I am able to paddle by a haul out area, a place seals congregate at low tide. At these times I have a chance to observe more than just the head. Though curious, seals are cautious creatures, I can never get too close before they slide into the water. Thus, it is nearly impossible to get to recognize any one seal by sight. Fortunately, seal researchers are making progress in understanding these enigmatic creatures.

(NOTE: seals are protected by federal laws and shouldn’t be approached closer than 50 yards or 150 feet)

Photo Identification

The life history of harbor seals is not well understood because, unlike humpback and orca whales, highly reliable photo identification methodology has been lacking. Photo identification has been used for decades in the research of the distribution, lifestyle, and migration patterns of whales and dolphins. Individual animals are identified by comparing photos to those stored in a database.

A number of years ago I wrote an article for Cetus magazine titled Wildlife studies in black and white: Photographic identification. I had been in the San Juan Islands of Washington State to learn about orca aka killer whales. The technique of photo identification for orca in this area was pioneered in the 1970s by Canadian researcher Dr. Michael Bigg who was studying the Southern Resident orca whale population found along the Canadian province of British Columbia and Washington state in the U.S. Individual orca are recognized by distinctive nicks, scars and shapes on their prominent dorsal fins. The dorsal fin characteristics are used in combination with the also distinctive whitish-grey pigmentation behind the fin, an area often referred to as the saddle patch. Individual animals are identified by comparing photos to those stored in a database. Bigg proved the reliability of this photo identification technique after he radio tagged an individual with two nicks carved in the rear edge of the dorsal fin. The nicks remained unchanged throughout the animal’s lifetime.

The longest and most detailed photo identification data sets in the world were taken of baleen whales in the Stellwagen National Marine Sanctuary, an 842-square-mile body of water located at the mouth of Massachusetts Bay, between Cape Cod and Cape Ann. With one baleen whale, the humpback, identification photos are taken of the underside of the fluke or tail. Individual individuals are identified by the unique patterns of the black and white pigmentation and scars on the fluke. This information is used to assign individual humpbacks a catalog number. Their identification photographs are retained in the Gulf of Maine Humpback Whale Catalog and the North Atlantic Humpback Whale Catalog, respectively archived by the Provincetown Center for Coastal Studies and the College of the Atlantic in Maine.

Facial Recognition Software

Accurate data on harbor seal populations is useful because these marine mammals serve as indicators of ecosystem health and knowing more about their behavior assists wildlife managers in making management decisions. Researches have found that photos of seal pelage is not trustworthy for a number of reasons including changes that occur as seals mature and during periods of molting. Other forms of monitoring techniques such as attaching tags have been used. These types of techniques are typically costly to do, time-consuming, and invasive.  

Facial recognition technology has proven to be a highly accurate tool in the surveillance and authentication of human faces. It also has been used with much success in primate research. This is why Krista Ingram, a professor of biology at Colgate University, thought it was worth a try. Although she acknowledged that “Seals are one step further away from the human face, so I wasn’t sure if this would work.”

The Colgate study involved dozens of harbor seals. They were photographed in Casco Bay, Maine in an area near where I kayak. A data base was created from the photos and machine learning software was created. The software nicknamed SealNet is designed to identify a seal’s face using facial features such as eyes and nose shape. More than 1700 photos of more than  400 individual seals were used to test the SealNet software. It proved to be 95% accurate. This level of accuracy in important for researchers like Ingram to understanding seal dispersal. “For mobile marine mammals that move around a lot and are hard to photograph in the water, we need to be able to identify individuals,” Ingram said.

The Need for Additional Data Collection and Value of Citizen Scientists

While SealNet has been heralded as a major breakthrough for work with harbor seals, less sophisticated photo identification methodology has been used by other groups working in the field for many years. A group known as Harbor Seals of Pacific Grove in Monterey County, California, has been photographing harbor seals for more than a decade. HSPG shares their work on their Facebook page. The group attributes their work to Bay Net, a group working in the Monterey Bay National Marine Sanctuary.

“We do it differently than the facial recognition software. We look at the spots and markings on the whole seal so that no matter which direction the seal is facing or what body part is showing, we have a good chance of identifying that seal,” said a spokesperson for Harbor Seals of Pacific Grove.

Many questions remain about wild populations of harbor seals; thus, there is a need for additional ecological and behavioral information. SealNet’s application can provide much new and useful information and its utility will likely extend beyond harbor seals. It could be extremely important in the study of more endangered marine mammal populations like the Mediterranean monk seal. A seal thought to be the world’s most endangered seal species.