Electronic monitoring (EM) is a technology with the potential to increase fisheries monitoring coverage, improving data quality and timeliness of data. This is particularly the case in the monitoring of bycatch. In many fisheries, bycatch events are rare and require high-levels of monitoring coverage (observer and/or video review), which can be time and cost-prohibitive. EM programs designed for bycatch monitoring should integrate artificial intelligence and machine learning (AI/ML) tool to efficiently detect rare events. There are a number of initiatives worldwide promoting aggregation of imagery into shared libraries to improve AI/ML development. To further our understanding of protected, endangered and threatened (PET) species bycatch, it is imperative to identify ways to aggregate and share AI/ML training datasets.
Few EM programs focus on bycatch monitoring, but those that do tend to use EM to audit self-reported catches (including logbooks). Programs with other focusses could improve bycatch estimates by integrating additional objectives as they mature. This is particularly important for PET species, where self-reporting is likely lower due to perceived higher risk of sanctions.
The objectives of this theme session are:
- to identify datasets and initiatives that support sharing of training datasets and promote the development of common image libraries to advance AI/ML
- to share progress from bycatch monitoring EM programs, including implementation practicalities, challenges and/or opportunities for further integration of data to improve fisheries management
We are seeking contributions regarding the following topics:
- examples of training datasets and shared image libraries across agencies and programs, their availability, governance, and potential to collect imagery by other sources
- examples of EM programs focusing on bycatch monitoring that have contributed to science, stock assessments, control, management, etc
- examples of how EM bycatch monitoring products can be provided to best suit stakeholders’ needs for improving fishing practices and minimising impacts.
