A Survey on Marine Debris Detection and Location Using Machine Learning Technologies

The large amounts of debris present in our ocean water poses a major threat to the flora and fauna of our marine ecosystem. The continuous increase in dumping of solid waste, and the very slow rate of degradation of most items, are together leading to a gradual increase in marine litter found in the ocean. While a large number of human-based campaigns have been proposed to tackle this issue, the sheer amount of litter present in the ocean results in their efforts to be insufficient. But the increasing development in the field of computer vision combined with the help of Autonomous Underwater Vehicle (AUV), presents itself to be a better solution to reduce the human complexities and automate the tedious task of decontaminating our ocean. This paper contains a literature review of the different Machine learning models using various neural architectures currently used to detect and locate various debris present in marine water (i.e., plastic, metals, glass, styrofoam, cigarette butts etc.).