Deep Learning based Automated Waste Segregation System based on degradability
Surendra Kumar Koganti, G Purnima, Pechetti Bhavana, Y Veera Raghava, Reghu Resmi · 2021 Second International Conference on Electronics and Sustainable Communication Systems (ICESC) · 2021
Rapidly increasing rate of consumption of resources by the growing population has led to a drastic spike in the amount of waste generated. Due to lack of effective management of waste at the initial stage of waste disposal from domestic as well as industrial sector causes segregation and recycling process to be challenging. A model is suggested in this paper which focuses on the software portion of an automatic waste segregation system where a camera and Raspberry Pi are mounted to detect and classify the individual waste item placed on a conveyor belt that carries the waste to the respective dustbin based on the classification done by Pi module. The software module consists of a Deep Learning algorithm called Single Shot Detector model with MobileNet as base network to classify the waste into biodegradable and non-biodegradable.