ARTIFICIAL INTELLIGENCE APPLIED TO MULTISPECTRAL IMAGERY FOR FLUVIAL MACROPLASTICS DETECTION

The research treats the detection of macroplastics in fluvial environments with multispectral images obtained by means of UAV proximity sensor. Further it will be developed an automatic classification methodology through Artificial Intelligence, machine learning and deep learning algorithms. The final goal of the project is to build a cheap methodology to be used for the periodic monitoring and create a digital georeferenced cartography to be updated and easily usable from local administrations and communities. Thanks to an efficient cleaning of riverfront, consequently it is possible to have an urban and social redevelopment of this fundamental area for the cities.