An Object Detection and Scaling Model for Plastic Waste Sorting

Padalkar, Abhishek Sunil (2021) An Object Detection and Scaling Model for Plastic Waste Sorting. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (5MB) | Preview
Preview	PDF (Configuration manual) Download (13MB) | Preview

Abstract

Plastic waste sorting in recycling industry involves mechanical and chemical separation of plastic into Polyethylene Terephthalate (PET), High-Density Polyethylene (HDPE), and Polypropylene (PP). Currently, Polystyrene (PS) and “Other” types of plastic are not sorted for recycling. Research has shown that higher recycling output is possible if most plastic types of waste are segregated at first. This research proposes an Object Detection framework solution to sort plastic waste. The framework combines an Object Detection and Scaling model (ODSM) and an Artificial Neural Network (ANN) model incorporating implicit features to detect f ive different types of plastics: PET, HDPE, PP, PS, and “Other”. The “WaDaBa” Plastic waste dataset is used for training purposes which consists of four thousand plastic waste images. This image classification dataset is pre-processed to object detection dataset, and pre-trained Scaled-Yolov4 and EfficientDet scaling models are applied on the plastic dataset. The results of eight trained models are presented in terms of accuracy, mean average precision(mAP), f1-measure for each plastic type, train time, inference time, and model size. This research demonstrates the potential of using the Scaled-Yolov4-CSP object detection model on higher resolution images to sort plastic waste in the recycling industry.

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software T Technology > T Technology (General) > Information Technology > Computer software G Geography. Anthropology. Recreation > GE Environmental Sciences > Environment
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Clara Chan
Date Deposited:	11 Dec 2021 11:43
Last Modified:	11 Dec 2021 11:43
URI:	https://norma.ncirl.ie/id/eprint/5206

Actions (login required)

View Item