Survey: Garbage collection and segmentation system using Mask-RCNN based Deep learning algorithms

Developing countries are experiencing a significant problem with waste pollution. There is well-known importance to recycling, both economically and ecologically, and the industry demands high efficiency. Because there are no benchmarks and widely accepted standards regarding the metrics and data used in current studies regarding automatic waste detection, they are hardly comparable. It is a relatively new field of research that has attracted a lot of attention recently. It has emerged as a powerful tool for revealing hidden patterns, determining correlations, or uncovering hidden patterns that traditional approaches cannot reveal. It summarizes and presents the results of previous studies on the presented datasets as well as the authors' own experiments. In this systematic literature review (SLR), 10 studies have been compiled from journals and conferences published between 2018 and 2022. In these studies, Deep Learning models were implemented and analyzed in various Solid Waste Management areas. Research studies were selected based on various criteria and quality assessment processes defined in this paper.