Imagine trying to fish for one specific type of fish but accidentally catching dolphins or turtles in your nets. This project uses smart cameras and AI to spot these animals automatically and track where they swim. By knowing exactly where the animals are and how the nets work, fishing boats can avoid the wrong areas and change their gear to stop accidental kills.
Industrial fisheries face financial and legal risks due to the accidental capture of protected species. Current monitoring is fragmented and relies on manual observation, which is expensive and often inaccurate.
Automated Machine Learning systems for bycatch detection and high-resolution predictive habitat maps for marine megafauna.
If you are a software provider dealing with inaccurate manual reporting of catches — this project developed automated Machine Learning (ML) systems that identify species-specific bycatch in high resolution. This allows for more precise data collection and automated reporting for industrial fleets.
If you are a fleet operator dealing with regulatory pressure and accidental species loss — this project developed a way to test mitigation measures and map risk zones. This helps you avoid high-risk areas and implement gear changes that reduce the loss of protected species.
If you are a consultant dealing with fragmented data for ocean zoning — this project developed predictive habitat mapping and tracking data for megafauna. This provides the scientific basis to design protected zones that actually align with animal movements.
Based on available project data, there is no information regarding the pricing or cost of the machine learning systems.
The project focuses on the East Central Atlantic Ocean, but the use of automated Machine Learning and electronic monitoring is designed for industrial fisheries, which could potentially be adapted to other regions.
Based on available project data, specific IP or licensing agreements are not mentioned; however, the project emphasizes open data management practices.
The project supports the goals of the UN BBNJ treaty and the EU’s Blue Economy by providing a scientific basis for political change and sustainable management of fisheries.
The project period runs from 2024-01-01 to 2027-12-31, suggesting that final validated results will be available by the end of 2027.
The consortium is heavily weighted toward research and academia, with 4 universities and 6 research organizations. However, it includes 13 partners across 5 countries, including 3 SMEs, providing a bridge between scientific discovery and industrial application. The 8% industry ratio suggests the project is primarily in the validation and testing phase rather than full commercial deployment.
Contact the Universitat de Barcelona for details on ML bycatch detection implementation.
Contact us to find partners for scaling these AI monitoring tools to other maritime regions.
