Use of High Quality UAV Based Data to Correct Satellite Derived DEMs

"The goal of this SBIR proposal is to develop and demonstrate a technique for improving the accuracy of DEMs generated from satellite imagery by using appropriate airborne imagery. It is anticipated that the positioning and pointing capabilities of futureUAVs will provide much more accurate geo-location of features on the ground than satellite systems and they, therefore, can provide the surrogate "ground truth" for correcting satellite data that has much larger area coverage. The algorithm will bedeveloped and exercised using IKONOS satellite-derived DEMs in combination with imagery data from the Arete Associates developed Airborne Remote Optical Spotlight System (AROSS), which a field-tested and robust EO system designed for research on tacticalUAV applications such as this one. In Phase 1, we will develop the algorithm for extracting highly accurate 3-D geodetic coordinates for multiple images collected from this UAV surrogate and develop the algorithm that uses these results to correct thesatellite data. In addition, we will evaluate the approach by applying the correction algorithm to an IKONOS imagery-derived DEM, using AROSS data and in situ ground truth collected during a recent experiment. Assuming success, phase 2 will developrobust software to automatically make these corrections of satellite data using UAV data, and perform a more complete evaluation of the accuracy on a much larger data set. Phase I will develop enha