Perpustakaan Digital ITB

Plastic waste that is difficult to decompose has become a global problem besides the ozone depletion and ocean acidification because it causes environmental damage. The problem of plastic waste also affected the coastal ecosystem area of Bali Island Indonesia, which has a distribution of plastic waste at a high to very high level in several locations. High distribution of plastic waste is caused by various activities on land and sea, therefore efforts are needed to reduce the impact of environmental damage. In this study, the identification of plastic waste in coastal ecosystem areas with medium resolution satellite imagery data is the main goal as an effort to reduce the impact caused by plastic waste through a low-cost monitoring program. Identification of plastic waste is carried out using two different analytical methods i.e. machine learning and object-based image analysis. Furthermore this study also aims to see the advantages and disadvantages of the analytical method used by utilizing previous research data as comparison. The novelty of this research lies in the use of analytical techniques that combine geographic information science technology and machine learning to identify plastic waste and develop a new ecosystem as an area of interest for research consisting of coastal, intertidal and mangrove areas. The results show that in general, geographic information science technology and machine learning can be used properly to identify plastic waste in the coastal ecosystem area of Bali Island so that it can assist in monitoring program, therefore this research can be used as an effort to prevent marine pollution, especially from activities on land, including marine debris and nutrient pollution contained in point 14.1 of SDGs.