Mapping riparian zone macro litter abundance using combination of optical and thermal sensor

A significant increase in the world's population will lead to an increase in consumption and, therefore, an increase in global waste. Various attempts have been made to monitor and map waste, but the proposed approaches are difficult and complicated, and they incur high costs. In this study, to overcome limitations in monitoring and mapping plastic waste, using combined optical and thermal sensors installed on drones is proposed. The study area is the riparian zone, or the zone around the river, where the accumulation of plastic waste at the mouth of the river eventually reaches the sea. The image data obtained were processed using machine learning methods to produce high accuracy and precision. To determine the effectiveness of the proposed method, an accuracy assessment was conducted. The results of this study indicate that the combination of optical and thermal sensors provides the best accuracy compared to using only single optical or thermal image data.