Case Study: Sea Cucumber Monitoring

Published On: January 12, 2023Categories: ,

With the support of the National Research Council of Canada (NRC) and the Atlantic Canada Opportunities Agency (ACOA), Marine Thinking was able to apply its proprietary artificial intelligence and machine learning algorithms to automatically recognize and detect sea cucumbers living along the ocean floor.

The team used remote operated vehicles (ROV) to navigate the ocean floor, collecting hours of optical footage across multiple trips.

Using portions of the captured footage, several large data sets were created with sea cucumbers manually identified and labelled for the machine learning algorithm. As part of the training process, image augmentation techniques were applied to optimize the learn quality. These techniques help us prevent data overfitting, increase prediction confidence, and improve the overall variability of the data models.

Once fully trained, the AI algorithm was able to successfully detect and identify the sea cucumbers during real-time playback of the footage.

To visualize the data collected, 3D maps were produced for each fishing area. Each map was populated with GPS coordinates and a visualized representation (location and density) of the projected total of sea cucumbers living in each area.

If you have project that would benefit from our AI object recognition and tracking technology, we’d love to hear from you.

Contact Us >

Project Details