Assessment of Waste Management through Mobile Edge Computing and Deep Learning

Abstract

Due to the random occurrences of street waste, city managers usually spend a lot of energy and money cleaning street garbage, which is a core duty in computer vision, with applications ranging across the process of smart city creation. Deep network solutions are frequently constrained by the amount of training data available as they become deeper and more complicated. With this in mind, Open CV or Google AI has made the Open Images dataset publicly available in order to drive breakthroughs in image analysis and interpretation. Open Images continues the legacy of PASCAL VOC, Image Net, and COCO, but on a much larger scale. As a result, visual street cleanliness assessment will be extremely vital in this project. Existing assessment methods, on the other hand, have several significant drawbacks, such as the lack of automation in the collecting of street waste data and the lack of real-time street cleanliness data. Finally, the findings are fed into a framework for calculating street cleanliness, which allows for the visualisation of street cleanliness. Cleanliness levels are maintained at a high level, making it easier for city managers to schedule clean-up crews.