Publication Date: 2020/10/20
Abstract: Solid waste categorization is challenging process but it is an important for recycle and disposal wastes in a proper way. Currently in Ethiopia solid waste are categorized manually as recyclable, combustible and compostable. We propose the machine learning approaching order to categorize the solid waste as recyclable, combustible and compostable through the machine. The scope of the study is to detect and categorize the solid waste. The solid waste collected from household, street sweeping, hotels, industries and other industries from Addis Ababa, Ethiopia. For experimental purpose we collected total 2445 images. Among these we found 650images are recyclable,865 images are combustible and 930 images are compostable category. The overall accuracy of our designed system is 89% and the model achieved 89%, 82% and 96%inrecyclable, combustible and compostable category respectively. The designed approach accuracy result is compared with manually identified categories and the average percentage error is 10.82% and the designed approach performs closer to the ground truth categorized manually
Keywords: Solid Waste, Digital Image Processing, Image Segmentation, Machine Learning, Artificial Neural Network, Object Classification
DOI: No DOI Available
PDF: https://ijirst.demo4.arinfotech.co/assets/upload/files/IJISRT20OCT202.pdf
REFERENCES