Pelagic vessels bring aboard very large catches of fish like mackerel and herring in a short time. Cameras on board record it, and those cameras produce far more video than any team of people can watch. Finding the few minutes that matter inside many hours of footage is slow, manual work. The Scottish Government wants advanced AI tools to support its science and cut down the time spent on review.
We are building computer vision models that find key events in the footage, such as when catch is being pumped aboard, and flag what deserves a closer look. The tools are tested on different vessel types and camera setups, so they hold up across the fleet.
Our AI finds the key moments in the footage, so scientists and reviewers can focus their time where their expertise matters most.
This project builds on years of electronic monitoring work with NOAA, Fisheries and Oceans Canada and fisheries in Alaska and Hawaii. It brings that experience from the Pacific to the North-East Atlantic, and it adds one of Europe's leading fishing nations to the list of governments that trust Ai.Fish with their fisheries video.
The project runs through February 2028. If your agency or fleet has more video than people to watch it, we would like to hear from you.




