Mr Celuxolo Dlamini (23) is well on his way to achieving his goal of a PhD by the age of 25 after being awarded a cum laude Master’s Degree in Environmental Science for using remotely sensed data from unmanned aerial vehicles (UAVs) or drones and machine learning techniques to predict maize yield.
Originally from Pongola, where he attended Landokwakhe High School, Dlamini was motivated to investigate the discipline of environmental science by two of his teachers, Mr BS Mdamba and Mr B Mabuza who had studied in the field before becoming teachers.
Inspired by what he saw at UKZN, and the example set by former students from his high school, UKZN was his only choice for tertiary studies. The oldest of seven siblings, he is the first to study to masters level and has a younger brother at UKZN and a sister at the Durban University of Technology.
Dlamini completed his undergraduate and honours studies on the Pietermaritzburg campus, seizing every opportunity to develop his personal and academic profile. He participated in student holistic development opportunities, conferences, field trips, leadership positions, contract employment, projects and short courses during his time as a student. A keen soccer player, he played in the campus soccer league, which helped him maintain his physical and mental health and balance the demands of his studies.
Having always aimed to progress to PhD level and beyond, Dlamini registered for his Masters in Environmental Science, focusing on precision agriculture. This was motivated by the shrinkage in the food basket that Dlamini observed because of multiple challenges, including poor soils, low capital among small-scale producers and climate change. He hoped to further the body of knowledge in this novel area of research to diversify food supply and production by improving yield and lowering production costs.
Despite not having funding when he began master’s studies, Dlamini was able to find work and accommodation as a residence assistant through UKZN’s Department of Student Residence Affairs and financial support from his supervisors, particularly Professor Onisimo Mutanga’s National Research Foundation – funded South African Research Chair (SARChI) in Land-Use Planning and Management.
His master’s research focused on the utility of UAVs equipped with smart sensors and machine learning methods for forecasting maize yield. A primary objective was systematically to review advancements in using UAV-remotely sensed data and machine learning algorithms to predict maize yield and identify potential challenges and opportunities in this domain.
The work led to the development of a paper under review for publication, with a second paper following, also under review, that used these insights to address the gaps identified and develop a deep learning model for predicting maize Above-Ground Biomass (AGB) by integrating UAV-remotely sensed data with field-measured biophysical and landscape variables in small-scale farming settings.
‘Timely and accurate maize yield predictions allow for the development of models that can precisely predict yield before harvesting which is useful for strategic evaluations, financial planning, efficient irrigation, and food production management,’ said Dlamini.
Achieving excellence in his master’s study was the result of Dlamini’s focus on dedication, passion and consistency. These are necessary, he explained, to commit fully to one’s studies and make the required sacrifices. His love for his research topic kept his mind active with ideas, even in the middle of the night, and he refused to rest on his laurels but continued to arrive consistently at the office, be available for further research opportunities and make the most of each day.
Dlamini has nurtured the dream of attaining his PhD by the age of 25 and has registered for his PhD studies, still under the supervision of his master’s supervisor Mutanga and co-supervisors Professor John Odindi and Dr Trylee Matongera, on the topic of UAV-Thermal Remote Sensing-Based Phenology Assessment and LIDAR-based Biomass Prediction of Sweet Potatoes Using Deep Learning. He hopes to go on to postdoctoral research to continue building his research, teaching and community engagement profile.
Dlamini thanked Matongera for his trust and all he taught him, and Mutanga and Odindi for their supervision and support. He also thanked his friends and family for their help, including his family’s care for his five-year-old son while he completed his studies.
Words: Christine Cuenod
Photograph: Supplied