Authors: Narcisa Pricope*, University of North Carolina Wilmington, Kerry Lynn Mapes, University of North Carolina Wilmington, Britton Baxley, University of North Carolina Wilmington, Kyle Woodward, University of North Carolina Wilmington
Topics: Remote Sensing, Applied Geography
Keywords: UAS, processing parameters, photogrammetry, remote sensing
Session Type: Paper
Start / End Time: 9:55 AM / 11:35 AM
Room: Harding, Marriott, Mezzanine Level
Presentation File: No File Uploaded
There is a growing demand for the collection of ultra-high spatial resolution imagery such as that which can be acquired using unmanned aerial systems (UASs). UASs are cost-effective on small scale and can fly at much lower altitudes thus yielding spatial resolutions not previously achievable with manned aircraft or satellites. The use of commercially available software for imagery processing has also become commonplace due to the relative ease at which imagery can be processed compared to traditional methods of aerial photogrammetry and the minimal technical knowledge required by users. Commercially available software such as AgiSoft Photoscan and Pix4D Pro Mapper are capable of generating the high-quality data that are in high demand for environmental remote sensing applications. While these software packages are user-friendly, they are somewhat of a “black box,” where the processes that are occurring are relatively unknown to the user. We seek to explore the implications of processing parameter decision-making on the desired outcomes. Some studies report what they consider to be the optimal parameters for processing between multiple software systems (Gross and Heumann, 2014), but there is still a need to quantitatively assess how the minute adjustment of just one or two variables over the entire available range affects final outputs. This research will quantitatively describe the effects of varying both keypoint image scale and image scale used in point cloud densification on the accuracy and quality of the resultant orthomosaics and digital surface models using multi-sensor imagery (RGB, multi-spectral and thermal.