A Deep Learning Based Multiclass Segregation of E-waste using Hardware Software Co-Simulation. Issue 1 (August 2021)

• Record Type: Journal Article
• Authors: Elangovan, S.; Sasikala, S.; Arun Kumar, S.; Bharathi, M.; Naveen Sangath', E.; Subashini, T.
• Is Part Of: Journal of physics. Volume 1997:Issue 1(2021)
• Publication Date: 2021-08
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1997/1/012039
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Today, the advancement in technology has potentially changed the lifestyle of all the people. Though this innovation is beneficial, it has quite adverse effects on both human health and environmental health. One of the main causes is 'E-Waste' produced by electronic gadgets. Globally, the usage of electronic gadgets has increased the quantity of "e-waste" or electronic waste and it has now grown a major problem. An unproper disposal of e-waste is now becoming an environmental and public health issue, as this kind of waste has become the most rapidly increasing segment of the municipal waste stream in the world. But this ever-increasing waste is very complex in nature and is also a rich source of metals such as Neodymium, Indium, Palladium, Tantalum, Platinum, Gold, Silver, Aluminium and Copper which can be recovered from these wastes and brought back into the production cycle and day to day utilization. Hence, there is a need of proper e-waste segregation and management to recover the precious materials from these kinds of wastes. In this project, a Deep learning model implemented using NVIDIA's Jetson Nano development kit has been proposed, to classify the waste components into two categories based Precious or Non – Precious metals present in the waste. The prototype model developed in turn segregates the waste with good accuracy and less time consumption.