Marine imagery has become a crucial non-destructive tool for studying changes to benthic ecosystems, with remarkable advancements in the number, resolution and spatial coverage of images due to improved platform design. It is also becoming a powerful tool for raising awareness about the importance of marine biodiversity conservation.
To extract quantitative data from these images, they must first be annotated, involving labelling captured objects of interest to derive physical, biological, and ecological information. Historically, this labelling has relied on human efforts. However, recent technological advances, particularly the integration of artificial intelligence (AI) into processing pipelines, is revolutionising the field, enhancing our ability to extract robust information and develop an unprecedented understanding of marine environments.
In this conference presentation, we will showcase how the 'Understanding Marine Imagery' (UMI) Subfacility of IMOS has implemented AI into SQUIDLE+ to enable national-level reporting. We will discuss current strengths and limitations of AI in developing these indicators. Additionally, we will highlight the potential of open marine imagery for public outreach, showcasing collaborations between UMI and Seamap Australia in creating immersive mapping content. This aims to engage the public and non-specialists in marine conservation, raise awareness about marine ecosystems, and promote sustainable management practices.