Biological Systems Engineering
First Advisor
Yeyin Shi
Second Advisor
Yufeng Ge
Third Advisor
James Schnable
Date of this Version
Summer 8-27-2023
Citation
Izere, P. 2023. PLANT HEIGHT ESTIMATION USING RTK-GNSS ENABLED UNMANNED AERIAL VEHICLE (UAV) PHOTOGRAMMETRY (Master's Thesis). Biological Systems Engineering--Dissertations, Theses, and Student Research. Department of Biological Systems Engineering, University of Nebraska, Lincoln.
Abstract
Using Unmanned Aerial Vehicle (UAV) based remote sensing technology to provide high-throughput plant phenotypic traits has shown a great potential of eliminating manual measurements. For example, the use of UAV imaging to estimate plant height is of great interest to plant breeders and agronomists. Height is linked to a plant’s physical growth, light competition, health, biomass, and yield; thus, its accurate estimation is crucial in breeding. Conventionally, accurate height estimation by UAV platforms relies heavily on the use of ground control points (GCPs) for geometric calibration due to the relatively low positioning accuracy of regular differential global navigation satellite system (D-GNSS). The process of setting up GCPs and measuring their accurate geo-coordinates in the field is time-consuming and labor-intensive. Hence, exploring the potential of Real-Time Kinematic GNSS (RTK GNSS) enabled UAV technology with centimeter level positioning accuracy to improve plant height estimation without GCPs is on demand. The goal of this study was to quantitatively evaluate the positioning accuracies of RTK GNSS enabled UAV technology and investigate its applications in plant height estimation. Specifically, we (1) conducted a thorough literature review on state-of-the-art remote sensing based plant height measurement technologies; then (2) quantitatively evaluated the horizontal and vertical positioning accuracies of commercial UAV systems in mosaicked maps with and without GCPs, one of them with regular D-GNSS and the rest two systems equipped with RTK GNSS technology; and finally (3) quantitively investigated the performance of the UAV systems on plant height estimation in a breeding setting, and evaluated the potential of RTK GNSS enabled UAV technology on accurate plant height estimation without GCPs. The study was carried out at three research farms of the University of Nebraska-Lincoln. Maize, Soybean, and Triticale crops were studied. Three methods were compared: (1) UAV technology with regular D-GNSS and geometrically calibrated with GCPs, (2) RTK GNSS enabled UAV technology and geometrically calibrated with GCPs, and (3) RTK GNSS enabled UAV technology and geometrically calibrated without GCPs, The results showed Method 3 with the lowest RMSE and highest coefficient of determination, ranging from an R-square of 0.808 to 0.966 for all locations, indicating high precision and strong correlation between the UAV-derived plant heights and manually measured ones. All three methods had a high comparative result in terms of vertical and horizontal positioning accuracies, R-square, and RMSE. However, Method 3 showed labor saving cost in comparison to the other three methods. Thus, it is the best alternative among the three. The study quantified the positioning accuracies of RTK GNSS-enabled UAV technology in mosaicked maps for small areas and demonstrated the capability of this technology for plant height estimation eliminating the need for GCPs. RTKGNSS-enabled UAV is a promising technology and its wide applications in various agricultural research and production will be on the rise.
Advisors: Yeyin Shi and Yufeng Ge
Comments
A THESIS Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Master of Science, Major: Mechanized System Management, Under the Supervision of Professors Yeyin Shi &Yufeng Ge, Lincoln, Nebraska, August 2023
Copyright © 2023 Pascal Izere