Plastics pose a considerable threat to the livelihood of the aquatic species and ecosystems as they can transport bio-accumulative harmful pollutants over long time intervals. Toxic smaller sized ocean plastics can be easily mistaken for food and hence can be transferred across aquatic vertebrates and invertebrates. As for the larger sized pieces like ghost nets, these have been found to be responsible for large amounts of accidental by-catch of pelagic species, seabirds, turtles or even cetaceans. In the last years, a number of initiatives have been started within the framework of cleaning, recycling and curbing the plastic pollution challenge. More recently, there was a G20 meeting in Germany that brought together world leaders and stakeholders linked by the Global Partnership on Marine Litter of United Nations Environment Programme, where they came up with an action plan on marine the litter. The G20 action plan echoed the need for innovative, affordable and sustainable approaches to better monitor, reduce and manage plastics in the aquatic environment.
Unmanned Aerial Systems (UAS) offer affordable 'above-the-head' monitoring allowing wide area coverage and very high geospatial resolution collection of information. Also, UAS can provide high – quality monitoring information and products about the coastal environment at the local scale, which can equip policy makers, stakeholders and citizens with a better understanding of marine plastic pollution as well as how to manage and reduce this problem. The aim of our vision is to develop the capability of detecting marine litter ( i.e. through statistics, plastic density maps, geolocation) using aerial data from drones.
Automatic detection and reporting are essential for such a procedure. We will use artificial intelligence (AI) algorithms for fast and accurate responses. A massive training data set with marine litter is required for the development of these algorithms, for which citizen science/crowdsourcing approaches will be applied. In particular, we will use crowdsourcing tools, for the visual interpretation of UAS imagery to detect marine litter. Plastic density maps will be produced using crowdsourcing tools and will be back reported to local NGOs or citizens thought a web-based platform.