 |
| Titel |
Comparison of SIFT and SURF based DEM extraction approaches on a GEOEYE-1 satellite stereo-pair |
| VerfasserIn |
Ioannis Daliakopoulos, Ioannis Tsanis |
| Konferenz |
EGU General Assembly 2014
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 16 (2014) |
| Datensatznummer |
250098880
|
| Publikation (Nr.) |
 EGU/EGU2014-14598.pdf |
|
|
|
|
|
| Zusammenfassung |
| A MATLAB module for Digital Elevation Model (DEM) extraction from Very High Resolution (VHR) satellite stereo-pair imagery is used to compare the efficiency of two well established feature detection and description algorithms. A procedure for parallel processing of cascading image tiles is used for handling the large datasets requirements of VHR satellite imagery. Scale-Invariant Feature Transform (SIFT) and Speeded Up Robust Features (SURF) algorithms are used to detect potentially tentative feature matches in the members of the stereo-pair. The resulting feature pairs are filtered using the RANdom SAmple Consensus (RANSAC) algorithm by using a variable distance threshold. Finally, tentative feature matches are converted to point cloud ground coordinates for DEM generation. A 0.5 m × 0.5 m Geoeye-1 stereo-pair acquired over an area of 25 km2 in the island of Crete, Greece is used as input for the module. The resulting 2 m × 2 m DEMs has superior detail over previously developed 2 m and 5 m DEMs that are used as reference, and yields a Root Mean Square Error (RMSE) of about 1 m compared to ground truth measurements. Results suggest that SURF’s superior runtime performance outweighs the slightly better feature quality attained with SIFT. |
|
|
|
|
|