Integration of Drone Multispectral Data for Mapping Winter Wheat and Weeds
Abstract
In precision agriculture (PA), it is essential to map the canopy cover of winter wheat and associated weeds’ cover to support crop monitoring and variable rate herbicide application. The objective of the study was to investigate the integration of multispectral drone data with Digital Surface Model (DSM) and vegetation indices (VI) for mapping winter wheat, chickweed, and hairy buttercup. 1) To map winter wheat, chickweed, and hairy buttercup using drone multispectral data, and 2) To analyze changes in mapping accuracies of winter wheat, chickweed, and hairy buttercup resulting from the integration of drone multispectral data with VIs and DSM. A Random Forest classifier based on machine learning was used to classify winter wheat, chickweed, and hairy buttercup. We discovered that the overall mapping accuracy of winter wheat and associated weeds' canopy cover improved by 6% when drone multispectral data was integrated with DSM and VIs, compared to drone multispectral data used alone. In addition, when drone data was integrated with DSM and VIs, respectively, the overall mapping accuracy increased by 3% and 0.3%. Improving the mapping of winter wheat, chickweed, and hairy buttercup will improve crop management, yield, and sustainability.
Subject Area
Geographic information science|Remote sensing|Sustainability|Environmental science
Recommended Citation
Ajibade Samuel Babatunde,
"Integration of Drone Multispectral Data for Mapping Winter Wheat and Weeds"
(2023).
ETD Collection for Tennessee State University.
Paper AAI30814161.
https://digitalscholarship.tnstate.edu/dissertations/AAI30814161