Smart City Surveillance: Automated Street Waste Detection Using Live Camera

Sarwar, Rehman (2024) Smart City Surveillance: Automated Street Waste Detection Using Live Camera. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (3MB) | Preview
Preview	PDF (Configuration Manual) Download (5MB) | Preview

Abstract

The study proposes an Automated Waste Source Separation system that utilizes the YOLOv8 models in detecting and classifying waste on the streets of modern cities. The system has two versions, YOLOv8s and YOLOv8l, both of which are built off more than six thousand images obtained from a diverse collection of cities. The images include organic waste, metal waste, glass waste, paper waste, and plastic waste. Such specific application aims to simplify low-complexity detection tasks while achieving reasonable performance with the YOLOv8s model with encoder of 0.9058 precision, 0.94279 recall and 0.96042 mean Average Precision score. On the other hand, the YOLOv8l model is reported to be less accurate in some scenes but provides better results in the task of detecting and classifying objects in cluttered scenes. Two models YOLOv8l and YOLOv8s are tested concerning precision, recall, F1 scores and mAP, whereby the concept of transfer learning was applied to make use of the pre-trained weights for minimization of training cycles. The system provides the highest performance results and hence is able to provide a great solution for a fast and effective waste segregation within the contemporary cities. The inclusion of YOLOv8 models in this system allows efficient waste detection which will ensure that the urban landscape is clean and free from dangerous diseases.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Abidi, Syed UNSPECIFIED
Uncontrolled Keywords:	Real-time waste segregation; YOLOv8small; YOLOv8large; waste detections
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > TD Environmental technology. Sanitary engineering G Geography. Anthropology. Recreation > GE Environmental Sciences > Environment H Social Sciences > HT Communities. Classes. Races > Urban Sociology > Urban Renewal
Divisions:	School of Computing > Master of Science in Artificial Intelligence
Depositing User:	Ciara O'Brien
Date Deposited:	20 Jun 2025 10:14
Last Modified:	20 Jun 2025 10:14
URI:	https://norma.ncirl.ie/id/eprint/7964

Actions (login required)

View Item