Waste Segregation into Biodegradable & Non-Biodegradable using Transfer Learning

Abstract

Garbage generation, inadequate waste collection, transportation, treatment, and disposal are serious environmental issues around the world. Because of rapid urbanisation and population growth, global annual waste generation is expected to increase from 2.01 billion tonnes to 3.4 billion tonnes over the next 30 years. In 2016, the world produced 242 million tonnes of plastic waste, accounting for 12% of all solid waste. The amount of waste generated in India is increasing remarkably that the current systems cannot cope with it due to the increase in urban population, and this impacts on the environment and public health. This paper proposes a smart bin concept using modern Artificial Intelligence techniques on a microcontroller-based platform. The primitive idea is to segregate waste after the waste is dumped, but the proposed system’s basic idea is to segregate the waste while being dumped. The software is designed in such a manner that it opens the corresponding bin on recognizing the type of waste using contemporary transfer learning methods. Once these smart bins are implemented on a larger scale, replacing the conventional bins today allows waste to be managed efficiently, thereby steering off dump yards.