DBPapers
DOI: 10.5593/SGEM2014/B23/S10.020

ENHANCEMENT OF THE VERTICAL ACCURACY OF LIDAR DATA WITH THE USE OF AUTOMATIC IMAGE MATCHING

W. Dominik
Wednesday 1 October 2014 by Libadmin2014

References: 14th International Multidisciplinary Scientific GeoConference SGEM 2014, www.sgem.org, SGEM2014 Conference Proceedings, ISBN 978-619-7105-12-4 / ISSN 1314-2704, June 19-25, 2014, Book 2, Vol. 3, 161-168 pp

ABSTRACT
The aim of this research was to verify whether it is possible to use the results of aerial image matching to increase the vertical accuracy of LIDAR data. The experiment consisted in generating a point cloud through automatic image matching, which then was compared with the LIDAR data. Image matching was carried out with the use of Image Station Automatic Elevation software which offers a feature-based matching algorithm. The main element of the study was to determine the optimal parameters of image matching. It was found that the most important parameters to control this process are the size of correlation window and the correlation coefficient threshold. After setting the optimal correlation parameters (correlation window 9 x 9 pixels, correlation coefficient threshold 99%), the image matching was carried out for the whole study area. The filtering of points lying above the ground from the point cloud obtained through image matching was conducted using the LIDAR data. The resulting data served as the basis for the determination of the difference in elevation between LIDAR and photogrammetric survey. The mean elevation discrepancy was determined in relation to the image matching results for each of the LIDAR strips individually. The elevation correction was then introduced to each of the LIDAR strips. The evaluation of the effectiveness of elevation correction of LIDAR data was conducted with regard to GNSS RTK field surveys. Statistical parameters showed that the DTM generated from the corrected LIDAR data is closer to the field surveys. It can be concluded that the image matching can provide very precise elevation data, which may be used for the correction of the LIDAR data.

Keywords: LIDAR, Matching, Correlation, Correction, DEM/DTM