The Repository @ St. Cloud State

Open Access Knowledge and Scholarship

Date of Award


Culminating Project Type




Degree Name

Geography - Geographic Information Science: M.S.


Geography and Planning


School of Public Affairs

First Advisor

Mikhail Blinnikov

Second Advisor

Michelle Stanton

Third Advisor

Jeoffrey Torguson

Fourth Advisor

Michner Bender

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Keywords and Subject Headings

UAV; Drones; Plastics; Waste; Environmental Monitoring; OBIA


There are growing concerns about the threats posed by plastics to human society and natural ecosystems. There is evidence of the harm presented to economies, public health and society. Although plastic pollution is an issue of great concern, low- and middle-income countries lack waste disposal services and this lead to disposal of waste including plastics into the environment. Monitoring presence of waste disposed into the environment is crucial for assessment of remedial measures . Traditional approach for identifying locations with plastic and waste accumulation in the environment involves field surveys, and drone technology is an emerging technology being applied for mapping the presence of plastics and waste in the environment. In this study, I have presented basic requirements for collecting data using Unmanned Aerial Vehicles (UAV) to map plastics and accumulation of domestic waste in the environment. For example, it was observed that a Ground Sampling Distance (GSD) of 2.51 cm is too coarse for mapping plastics of size less than 10 cm. Additionally, the study has also utilized random forest as a machine learning algorithm to classify and identify plastics and waste piles from UAV-derived imagery in a densely populated area of Blantyre, Malawi. The random forest predictions show high performance compared to prior studies for both waste piles (Precision: 0.9048, Recall: 0.95, and F-score: 0.9268) and plastics detection (Precision: 0.8905, Recall: 0.9421, and F-score: 0.9156). With the reported accuracies, UAV imagery can be employed to guide environmental policy implementation by helping in monitoring the effectiveness of policies that have been set to mitigate and address problems such as open waste dumping.


I wish to express my sincere gratitude to all individuals and institutions that helped me to bring this project into life.

Firstly, great thanks to my advisor Dr. Mikhail Blinnikov, for providing me guidance in developing this project. Thanks to Dr. Michelle Stanton for providing me with mentorship support. The same also goes to the other members of my thesis committee, other faculty members at the Geography and Planning department of the School of Public Affairs at SCSU, specifically Mr. Thomas Oien for assisting me in collecting UAV data for my experiment.

I am also thankful to Rosheen Mthawanji, Chifuniro Baluwa, and Taonga Mwapasa for assisting me in collecting data to support my experiment in Malawi. I also thank Sustainable Plastics Attitudes to Benefit Communities and their Environments (SPACES) for providing funding to support data collection, and GLOBHE, a private company that rendered the service to collect UAV imagery across the entire Ndirande neighborhood.

Finally, I am very grateful to the Fulbright Program for sponsoring my Master’s studies. In many ways, my experience in Minnesota has been intellectually rewarding.