Theme Issue: Integration of Geodata and Imagery for Automated Refinement and Update of Spatial Databases

摘要

Photogrammetric and remote sensing imagery is since a long time considered as the primary data source for topographic mapping. Traditionally, the sequence of operations was well defined: the images were captured, relevant features were manually extracted and plotted using stereo compilation or digitised in orthophotos, and finally these features were cartographically post-processed to produce a line map as the final result. With the availability of digital imagery, laser scanner data and geo-spatial databases, all stored in computer readable form, two important trends have emerged: (a) based on image analysis techniques feature extraction is becoming more and more automatic. Automatic and semi-automatic feature extraction has been a focus of research for a few decades, and results are now starting to enter into the commercial market. (b) The linear chain of processes has given way to a cycle of steps. Features already contained in the database are superimposed onto the imagery to determine, whether the landscape has changed and thus an update is necessary. At the same time, a refinement of the existing features in terms of an increased degree of detail, better geometric accuracy, further attributes and possibly also the third dimension is being carried out. In addition, the image data have acquired a new role, as orthophotos have come to constitute a nearly mandatory layer in topographic Geographic Information Systems (GIS), and for photo-realistic and possibly dynamic three-dimensional visualisation.