Bulletin des sciences géographiques de l'INCT
Volume 12, Numéro 2, Pages 18-26
2008-10-31
Automated Updating Of Building Data Bases From Digital Surface Models And Multi-spectral Images: Potential And Limitations
Authors : Rottensteiner F .
Abstract
A method for automatic updaling of building data bases from Digital Surface Models (DSM) and a normalised difference vegetation index is evaluated. The DSM can be generated from Airbome laserscanner (ALS) data or by image matching techniques. Buildings are detected automatically from the Ùtput data.The buildùtg detection results are compared to an existing building data base, and changes between the existing data base and the new data set are determined. Buildings and building parts are classified as being confirmed, changed, new, or demolished. Change detection considers the fact that the original data and the building detection results can have a different topology and that small differences between the data from the two epochs might be caused by generalisation errors, by a misalignment of the data, or by insufficient sensor resolution. The perfonnance of the algorithm is analysee! using DSMs generated both from ALS data and by image matching. The evaluation shows the different properties of these data for buildùtg change detection and also some of the limitations of the method. If the accuracy requirements for the building outlines are not vety high, the automatic updating process can be automated, provided that high-quality DSMs are used. In a semi-automatic environment the amount of human itteraclion for updating building da ta bases can be reduced by 40%-60%.
Keywords
Change Detection, Digital Surface Moclels, Data Fusion
Les articles similaires
Elfoul Lantri
.
pages 19-38.
Zerbo I.
.
Zoungrana M.
.
Seré A.d.
.
Ouedraogo F.
.
Sam R.
.
Zouma B.
.
Zougmoré F.
.
pages 517-532.
Ly I.
.
Lemrabott O.h.
.
Dieng B.
.
Gaye I.
.
Gueye S.
.
Diouf M.s.
.
Sissoko G.
.
pages 187-206.
Boukhentache Slimane
.
pages 12-29.
Chelghoum Nadjim
.
Zeitouni Karine
.
pages 114-130.
