A pipeline for Solid Domestic Waste classification using Computer Vision

Abstract

This work aims to build and analyze a pipeline for solid domestic waste classification. The first steps that were carried out for this were to divide into three main lines that work together to achieve the pipeline. Each line used different sub-approaches to deep learning, relying on both the literature and the advisors, but without neglecting the binary classification work previously carried out. Additionally, a CRISP-DM methodology is taken into account to carry out the work without taking apart the mathematical concepts behind each computationally implemented method, and it is specified what are the intentions of the authors for the near future and their conclusions.

Image Processing Using Morphology on Support Vector Machine Classification Model for Waste Image

“…These findings demonstrate the effectiveness of using the CNN algorithm with Multilayer Hybrid feature extraction for waste classification through image processing. In 2021, Gomez et al [16] conducted a study on waste image processing, evaluating multiple methods to identify the most accurate and appropriate approach for waste classification. The findings suggest that the SVM method is more effective in accurately classifying waste than other methods evaluated in this research.…”