This study explores land cover classification methods using high-resolution UAV imagery collected at Lakeside Farm, Oregon in July 2024, with a focus on mapping land cover types. The goal is to assess the effectiveness of multiple supervised classification techniques: K-Nearest Neighbor, Maximum Likelihood, Random Trees, and Support Vector Machine. Object-based and pixel-based classification methods were applied to RGB imagery collected using the Parrot ANAFI drone. The study area provided a unique array of land cover types, including varied vegetation, unvegetated areas, water, and wetlands, which help provide insight as to the advantages and disadvantages of the different classification methods when dealing with each land cover type. This research provides insights into the optimal conditions and methods for UAV data integration in agricultural land cover mapping, contributing to enhanced precision agriculture practices.
