Drone and Machine learning (artificial intelligence) to combat plastic pollution in rivers
Live Session 5: May 27 14:00-16:00 (GMT+7)
Recently a novel approach such as using a combination of unmanned aerial vehicle (UAV) surveys and image processing based on machine learning is being applied for assessment and monitoring for plastic pollution. Many activities have been carried out to study plastic pollution situation which is mostly based on in-situ manual observations like the plastic waste amount and composition survey. The survey has some challenging issues including time-consuming, the requirement of larger human resources, and the risk for human errors. The CounterMEASURE project has conducted the trial for the purpose to detect mechanically plastic product type and amount from the aerial image of the riverbed taken by using a drone.
The session shall provide the possibilities and challenges of the new tool and approach (drone and machine learning) to apply for the plastic leakage monitoring and assessment.
The session shall cover the following topics:
The use case of the drone survey with machine learning as a use case
The lessons learnt from the trial of CounterMEASURE project
Recent study result for plastic debris volumes estimation by using UAV and image processing based on deep learning
What to Learn
The participants will learn the current status of use cases of a combination of UAV surveys and image processing based on machine learning for plastic pollution as a tool for plastic leakage assessment and monitoring. In addition, the session can contribute to add new options and insights for participants to tackle the plastic pollution
Click photo to see the bio
Registration (Open Webex page)
Current trend of drone and UAV for plastic detection (Presentation Slide)
Aerial Image Analysis Survey in Mekong River (Presentation Slide)
Aerial Image Analysis Survey in Mekong River (Survey Report)
Kako, S., Morita, S., & Taneda, T. (2020) Estimation of plastic marine debris volumes on beaches using unmanned aerial vehicles and image processing based on deep learning. Marine Pollution Bulletin, 155.
Live Session Record