The rapid decline of biodiversity in coastal ecosystems presents a significant challenge to ecological stability. This study aims to develop an Artificial Intelligence (AI) driven framework to enhance biodiversity monitoring, specifically focusing on canopy-forming macroalgae in the Tuscan Archipelago. Leveraging emerging imaging technologies like satellite imagery, underwater cameras, and sensor networks, the study will employ deep learning-based computer vision (CV) for object recognition and percentage coverage estimation. This approach will facilitate improved species identification, assessment of habitat fragmentation impacts, and analysis of ecological patterns.

This methodology will involve training CV models with photo-quadrant images from within and around the study area. By integrating multi-source data, the study seeks to enhance real-time biodiversity monitoring. Model performance will be evaluated using standard metrics, including Precision, Recall, and F-1 score.

Anticipated outcomes will include improved species identification accuracy, enhanced understanding of the role of macroalgae ecosystem stability, and the development of a comprehensive biodiversity dataset. Integrating satellite imagery with underwater camera data will enable high spatiotemporal resolution monitoring of biodiversity trends. This research is expected to contribute to scalable and cost-effective monitoring tools, support ecosystem-based management strategies, and advance ecological research by demonstrating the potential of AI-driven emerging imaging technologies in coastal ecosystem conservation.