CHE01 - AI-Based Underwater Fish Detection and Tracking for Spatio-Temporal Marine Monitoring

Understanding the movement and spatial distribution of fish is fundamental to assessing ecosystem dynamics and supporting sustainable marine resource management. This challenge aims to develop an AI-based framework capable of automatically detecting and tracking fish in underwater video sequences acquired under varying visibility conditions. By integrating computer vision and multi-object tracking techniques, the proposed solution will transform underwater imagery into valuable spatio-temporal information, enabling the analysis of fish trajectories and movement patterns.

The resulting prototype can support applications in fisheries management, marine biodiversity studies, aquaculture, and ocean observation systems.

The challenge promotes the development of intelligent and non-invasive monitoring tools that convert underwater video data into actionable insights for marine science and conservation.