A machine learning algorithm was developed to analyse underwater videos and to detect the presence of macroalgae. Three habitat-forming algae Ecklonia radiata, Lessonia variegate, and Carpophyllum spp. were successfully identified. The machine learning models can be readily applied to ongoing monitoring programmes to rapidly determine and map the distributions of key macroalgal indicator species along coastlines. Monitoring data are critical to documenting changes in our coastal communities.
AEBR 263 Automated detection of large brown macroalgae using machine learning algorithms—a case study from Island Bay, Wellington
Report - Aquatic Environment and Biodiversity (AEBR)