Mapping waste piles in an urban environment using ground surveys, manual digitization of drone imagery, and object based image classification approach.

Kalonde, Patrick, Mwapasa, Taonga, Mthawanji, Rosheen, Chidziwisano, Kondwani, Morse, Tracy, Torguson, Jeffrey S, Jones, Chris ORCID: https://orcid.org/0000-0002-6504-6224, Quilliam, Richard S, Feasey, Nicholas ORCID: https://orcid.org/0000-0003-4041-1405, Henrion, Marc, Stanton, Michelle ORCID: https://orcid.org/0000-0002-1754-4894 and Blinnikov, Mikhail S (2025) 'Mapping waste piles in an urban environment using ground surveys, manual digitization of drone imagery, and object based image classification approach.'. Environmental Monitoring and Assessment, Vol 197, Issue 4, p. 374.

Text ManuscriptWasteMapping_kalonde.pdf - Accepted Version Available under License Creative Commons Attribution. Download (2MB)

Abstract

There is wide recognition of the threats posed by the open dumping of waste in the environment. However, tools to surveil interventions for reducing this practice are poorly developed. This study explores the use of drone imagery for environmental surveillance. Drone images of waste piles were captured in a densely populated residential neighborhood in the Republic of Malawi. Images were processed using the Structure for Motion (SfM) technique and partitioned into segments using Orfeo Toolbox mounted in QGIS software. A total of 509 segments were manually labeled to generate data for training and testing a series of classification models. Four supervised classification algorithms (Random Forest, Artificial Neural Network, Naïve Bayes, and Support Vector Machine) were trained, and their performances were assessed regarding precision, recall, and F-1 score. Ground surveys were also conducted to map waste piles using a Global Positioning System (GPS) receiver and determine the physical composition of materials on the waste pile surface. Differences were observed between the field survey done by community-led physical mapping of waste piles and drone mapping. Drone mapping identified more waste piles than field surveys, and the spatial extent of waste piles was computed for each waste pile. The binary Support Vector Machine model predictions were the highest performing, with a precision of 0.98, recall of 0.99, and F1-score of 0.98. Drone mapping enabled the identification of waste piles in areas that cannot be accessed during ground surveys and further allowed the quantification of the total land surface area covered by waste piles. Drone imagery-based surveillance of waste piles thus has the potential to guide environmental waste policy, offer solutions for permanent monitoring, and evaluate waste reduction interventions.

Item Type:	Article
Subjects:	W General Medicine. Health Professions > W 82 Biomedical technology (General) WA Public Health > Waste > WA 778 Waste products. Waste disposal
Faculty: Department:	Biological Sciences > Vector Biology Department Clinical Sciences & International Health > Clinical Sciences Department
Digital Object Identifer (DOI):	https://doi.org/10.1007/s10661-025-13675-6
SWORD Depositor:	JISC Pubrouter
Depositing User:	JISC Pubrouter
Date Deposited:	10 Apr 2025 10:28
Last Modified:	10 Apr 2025 10:28
URI:	https://archive.lstmed.ac.uk/id/eprint/26439

Statistics

View details

Actions (login required)

Edit Item