Using Uncrewed Aircraft Systems to Enhance Marine Debris Detection and Removal

Oregon State University, in collaboration with the NOAA National Centers for Coastal Ocean Science and NOAA Marine Debris Program are using uncrewed aerial systems (UAS) imagery, machine learning, and advanced sensors to detect and identify marine debris on shorelines.

Type of Project: Research

Project Dates: June 2020 - May 2022

Who is Involved?
Researchers from the NOAA Marine Debris Program, National Centers for Coastal Ocean Science, Oregon State University College of Engineering, and Genwest Systems, Inc. are using innovative approaches to detect and identify marine debris on shorelines in order to support and prioritize future marine debris removal efforts. The team will use emerging UAS sensors and platforms to collect imagery and train machine learning algorithms to identify different types of marine debris.

What is the project and why is it important?
Our ocean, waterways, and Great Lakes are polluted with a variety of marine debris, ranging from soda cans and plastic bags to derelict fishing gear and abandoned vessels. Locating and identifying marine debris quickly and accurately is key for cleaning up everyday debris and responding to debris created as a result of storms and disasters. Uncrewed aerial systems (UAS) can be used to collect imagery in difficult to reach or sensitive areas, or following a storm, to locate and measure debris introduction and impacts. However, with rapidly evolving technology and a wide range of systems and methods for collection and analysis of data, it can be difficult to consistently identify marine debris and compare the data gathered using this technology.

This project aims to create an accurate, low-cost, and easy-to-use marine debris detection system and develop standard operating procedures to improve the use of imagery collected through UAS sensors and platforms. Using this technology, researchers will collect imagery of shorelines in Texas and Oregon and develop heat maps that show the amount of marine debris present. They will also develop, train, and test machine learning to automatically detect and identify marine debris items in an image. The results of the project will be compiled into standard operating procedures, which can be used by the NOAA Marine Debris Program to plan and improve future UAS operations that guide cleanup efforts. 

Learn more about this project on the National Centers for Coastal Ocean Science website.