ARS | Publication request: A procedure for orthorectification of sub-decimeter resolution imagery obtained with an unmanned aerial vehicle (UAV)

Submitted to: American Society for Photogrammetry and Remote Sensing Proceedings
Publication Type: Proceedings
Publication Acceptance Date: February 29, 2008
Publication Date: April 28, 2008
Citation: Laliberte, A., Winters, C., Rango, A. 2008. A procedure for orthorectification of sub-decimeter resolution imagery obtained with an unmanned aerial vehicle (UAV) [abstract]. American Society for Photogrammetry and Remote Sensing Annual Conference, April 28 - May 2, 2008, Portland, OR.

Technical Abstract: Digital aerial photography acquired with unmanned aerial vehicles (UAVs) has great value for resource management due to the flexibility and relatively low cost for image acquisition, and very high resolution imagery (5 cm) which allows for mapping bare soil and vegetation types, structure and patterns in great detail. While image acquisition is relatively straightforward, the creation of orthorectified, GIS-ready image mosaics presents multiple challenges. Those include relatively small image footprints, image distortion due to the use of low-cost digital cameras, difficulty in locating ground control points and in automatic generation of tie points, and relatively large errors in exterior orientation (camera position and attitude information from the UAV’s GPS/IMU). We developed an automated procedure to improve the accuracy of the exterior orientation by matching the UAV images to an orthorectified reference image. Using the UAV reported exterior orientation and camera geometry, combined with the reference image and DEM, the algorithm simulates image acquisition and then computes the covariance between camera image and simulated image pixels. With this evaluation function, a heuristic search algorithm finds successive improvements to the external orientation, ultimately producing a corrected exterior orientation that allows orthorectification with minimal input of tie points and/or ground control points. The RMS error for a 5-cm resolution, 257-image mosaic was 48 cm. Cost and turnaround time for production of orthorectified mosaics from UAV imagery are considerable reduced due to less time and money spent on ground control point and manual tie point collection.