Mapping Marine Debris

Proposed in 2018 by a group of Hawaiian researchers, Mapping Marine Debris is a project that seeks to automate the image analyses of coastal pollution. After identifying a clear need for a better understanding of the amount, composition, and location of coastal pollution, a hybrid model was devised, leveraging the powers of citizen science and machine learning. The machine learning component revolves around the automation of images, including aerial photographs identifying debris densities and regional hotspots, ground-level photos taken by smartphones before and after cleanups, and drone photos quantifying microplastics. The machine learning algorithms build upon previous mapping efforts of the main Hawaiian Islands and provide precise measurements of quantity, location, type, and size of large debris in a variety of shoreline locations. Citizen scientists play a vital role in facilitating identification efforts (ie: providing photographs). The researchers are therefore also developing online platforms to engage concerned citizens and provide community organizers with interactive debris reports. As of July 2020, the project’s last update related to marketing assistance, signalling that the algorithms and online platforms have been successfully established. Despite this, the success of the project has gone unreported. The development of such platforms and algorithms could benefit a wide range of geographic areas, and will hopefully be expanded beyond Hawaii.

Moy, K., Neilson, B., Chung, A., Meadows, A., Castrence, M., Ambagis, S., & Davidson, K. (2018). “Mapping coastal marine debris using aerial imagery and spatial analysis“.

Categories

Artificial Intelligence, Citizen Science, Data, Ecological Monitoring, Monitoring, Pollution, Regulation, Visual Technologies