ABSTRACT

Unmanned aerial vehicles (UAVs), or drones, enhance the spatial and temporal resolution of data collected for natural resource management endeavors. Although user control of image acquisition is increased when collecting digital imagery with a UAV compared to traditional remote sensing methods, image acquisition over homogeneous landscape areas, such as even-aged forest stands, can result in poor-quality or incomplete orthophoto mosaics. To increase the usability of UAVs over homogeneous forested environments, pre-flight parameters, including altitude, overlap, and flight mission patterns, were assessed for their effects on orthophoto mosaic completeness, i.e., the percentage of images successfully tied together over the amount of holes/null data in an orthophoto mosaic. Imagery was collected over study sites in Nacogdoches County, Texas, and was processed into orthophoto mosaics with PIX4Dmapper photogrammetry software. The completeness quantification was computed using ArcGIS Pro and ERDAS IMAGINE 2022 geospatial software. Following imagery processing, completeness data were used to create a prediction model in R Studio to visualize the effects of pre-flight parameters and levels of canopy closure on orthophoto mosaic completeness. With the current model, UAV pilots should choose double-pass flight configurations flown at higher altitudes, specifically 200 ft or 400 ft above ground level, to generate orthophoto mosaics with higher completeness over forested areas. However, the prediction model will be revised as more observations are added to the completeness dataset.