Hammerhead Shark Photo ID Project

Turtle Island Restoration Network has launched a research project to determine if a computer algorithm can do what the human eye can’t— recognize individual sharks.

Studying wildlife populations underwater, especially of highly migratory species like hammerhead sharks, has been problematic for scientists and has prevented basic understanding of population dynamics necessary and critical to the conservation of shark species around the world.

TIRN has teamed up with Charles Stewart, Professor and Chair Computer Science Department at Rensselaer Polytechnic, and Jason Holmberg from Wild Me to create the Hammerhead Photo ID project using computer vision and deep convolutional neural networks.

If successful, the Hammerhead photo ID project will transform our knowledge of marine species and improve our understanding of population size, longevity, site residency, and movement of sharks. Furthermore, using historical photos, we will able to look back in time and estimate shark populations of the past so we can better understand the declines that make these species endangered today.

Updates