Waste Profiling and Analysis using Machine Learning

A large amount of solid waste is generated in the urban areas, and these wastes contain different types of substances like organic wastes, paper, plastic, metal, glass, etc. which needs to be treated separately for efficient waste management. This indicates that the wastes which are dumped all together need to be separated into categories before treating them. To support this process, Government introduced the concept of wet and dry wastes for civilians to dump wastes accordingly. If these norms are followed strictly, a huge amount of budget for waste segregation will be saved, which can be used for further waste treatment. Keeping this in view, this paper has proposed the system which can classify the waste as dry waste or wet waste based solely on the image of the waste taken. Focusing on simplicity, it is intended to propose an application which will only be required by the civic bodies to upload the captured images of garbage bins and sent to the system to analyze whether the garbage is wet, dry or mixed. The detection of contents of the garbage is the crucial aspect which will be done using machine learning. This idea can contribute in the near future to help analyze the waste disposal habits of people in different locations. This analysis can then be used to create awareness in required locations and help improve the waste disposal habits.