Publicly available datasets, such as images, videos, or time series, can be annotated to train machine learning and artificial intelligence (AI) models to support marine energy monitoring systems. High-quality and validated annotations are essential to produce accurate and reliable models, but the process is time-consuming and resource intensive.  

The Data Annotation for Marine Monitoring (DAMM) project aims to advance the application and efficiency of environmental monitoring through the development of annotated datasets for AI-based data processing and analysis algorithms. Ultimately, the DAMM project seeks to provide the marine energy industry with priority datasets and leverage data already collected from Triton and similar efforts. A pilot effort will demonstrate the transferability of data to new marine energy sites while also highlighting the limitations when applying these datasets in diverse environmental contexts.  

Another opportunity through the DAMM project is to generate standards and metrics to evaluate annotation quality. By establishing these guidelines and tools, DAMM can facilitate collaboration, maintain data integrity, and support the development of robust machine learning models for marine energy monitoring.