Automatic DTM generation from spaceborne KFA 1000 images - a contribution to the establishment of a digital landscape model

One major task within the Digital Landscape Model for Europe project of the European Organization for Experimental Photogrammetric Research (OEEPE) is the description of the shape of the terrain surface. This description will be obtained by manually controlled or automatic exploitation of stereoscopic image data in order to get an impression of the performances of the different procedures involved. A process for an automatic evaluation of digital image data is presented, resulting in a description of the surface by means of a digital terrain model. The procedure uses an image matching process based on a correlation procedure, which takes the spatial structure of the terrain surface into account allowing for the rectification of the image data. The surface is sampled by a regular grid of single points with constant step width. Scale-reduced spaceborne Russian KFA 1000 images (scale approximately 1:275000) have been used as image data. The images are digitized with a resolution of 50 /spl mu/m. The results of the point determination are presented and analyzed with respect to accuracy and morphological correctness. A high quality digital terrain model could be used as reference information.<
>