A Cost-Effective Smart E-Bin System for Garbage Management Using Convolutional Neural Network

• Published in: 2021 International Conference on System, Computation, Automation and Networking (ICSCAN) pp. 1 - 6
• Main Authors: A, Saranya, Bhambri, Mukul, Ganesan, Vinothkumar
• Format: Conference Proceeding
• Language: English
• Published: IEEE 30.07.2021
• Subjects: Classification; Classification algorithms; Convolutional Neural Network; Cost model; E-bin; Organizations; Real-time systems; Roads; Sensors; Smart Cities; Training; Urban areas; Waste Management
• DOI: 10.1109/ICSCAN53069.2021.9526547

The world has accumulated an enormous amount of waste that the ways for collecting and disposing of it are now out of human reach. The race is struggling to collect the waste and get it removed as soon as possible. Not only the non-biodegradable waste is creating issues but also the biowaste due to the lack of resources to collect them. Indian roads, sewages, even obsolete areas are plundered with plastic bottles, polythene bags, litters, etc. This project includes a cost-effective bin for small-scale purposes which is implemented in houses. The waste container (E- bin) consists of two divisions, one for the bio-degradable waste and the other for the non-biodegradable waste. The Proposed E-Bin consists of a sensing lid that detects if the bin is to be opened for the nearby person or not and waste type. It has two LED displays, one for the amount of waste collected and the other for how much maximum cost the waste could be sold to different organizations who can utilize the waste effectively. For waste classification Convolution Neural Network (CNN) is used to identify whether the correct type of waste is deposited to the correct bin. The accuracy of the waste classification algorithm is 96%. This proposed methodology is an initiative to encourage people to deposit and effectively dispose of their waste. Also, they are getting some reward for their waste.