Dolphin identification gets algorithm boost

Local marine biologists are part of a massive international collaboration that saw 35 humpback dolphin researchers from eight countries gather to share more than 1 200 photo-ID images of almost 300 distinctive humpback dolphins

Dave Savides

June 11, 2020

THE arduous task of identifying individual humpback dolphins through examining photos of their dorsal fins and comparing them to photo-ID images stored on a database could soon be improved through an ‘automated’ process.

And it’s an important gain since correctly identifying individuals will greatly assist studies of dolphin movements, population dynamics and conservation biology.

ALSO READ: Exercise your pet and help the SPCA

Indian Ocean humpback dolphins can be found from the southern tip of Africa to the southern tip of India in a very narrow strip of shallow coastal water.

They are endangered because their habitat is becoming degraded and they are killed by humans, usually unintentionally.

Their extinction rate worries dolphin researchers who are working to understand and change the situation.

Individuals like Zipper, La Linea and Venus (pictured), photographed at Richards Bay, can yield information about this endangered species
PHOTOS: Brett Atkins

Local marine biologists are part of a massive international collaboration that saw 35 humpback dolphin researchers from eight countries (South Africa, Madagascar, Tanzania, Kenya, Abu Dhabi, Dubai, Iran and India), gather to share more than 1 200 photo-ID images of almost 300 distinctive humpback dolphins.

South Africa’s SouSA Consortium, a group of researchers who study humpback dolphins at various sites along our coastline, submitted 569 of those images (141 distinctive humpback dolphins).

Photographs of the dolphins’ dorsal fins show nicks and notches, scars and colour patterns on and around the fin, allowing biologists to recognise distinctive individuals.

The process of photo-ID is time-consuming and this is where the exciting innovation comes in.

Recently, photo-ID of various species, including zebras, has been automated.

ALSO READ: New seahorse species discovered in Sodwana Bay

An automation front-runner is WildMe, a non-profit, wildlife conservation organisation, uniquely comprised of software and machine learning engineers.

They have been working with leading marine mammal biologists from the Indian Ocean Network for Cetacean Research (Indocet) and the Arabian Sea Whale Network to develop Flukebook, a platform to host and match the ID photos of humpback whales, bottlenose dolphins and whale sharks, using algorithms that examine the distinctive features to assess similarity and identify if that particular individual has been photographed before.

Adding endangered humpback dolphins to the algorithm-based identification system could go a long way to helping us understand them and ensuring their survival.

The International Whaling Commission has welcomed these efforts, recognising humpback dolphins’ dire situation and acknowledging the critical role that identifying individuals plays in conservation biology.

They encourage funding agencies and individuals to provide support for this project.