Integration of image analysis and GIS
Emmanuel P. Baltsavias, Michael A. Hahn · Repository for Publications and Research Data (ETH Zurich) · 1999
Photogrammetry and remote sensing have proven their efficiency for spatial data collection in many ways.Interactive mapping at digital workstations is performed by skilled operators, which guarantees excellent quality in particular of the geometric data.In this way, worldwide acquisition of a large number of national GIS databases has been supported and still a lot of production effort is devoted to this task.In the field of image analysis, it has become evident that algorithms for scene interpretation and 3D reconstruction of topographic objects, which rely on a single data source, cannot function efficiently.Research in two directions promises to be more successful.Multiple largely complementary sensor data like range data from laser scanners or SAR and panchromatic or multispectral aerial images have been used to achieve robustness and better performance in image analysis.On the other hand, given GIS databases, e.g.layers from topographic maps, can be considered as virtual sensor data which contain geometric information together with its explicitly given semantics.In this case, image analysis aims at supplementing missing information, e.g. the extraction of the third dimension for 2D databases.A second goal, which is expected to become more important in future, is the revision and update of existing GIS databases.In this paper, we review recent developments in the overlapping area of image analysis and GIS.On the data side, we focus on different sensor data in conjunction with GIS databases.Analysis of these data addresses almost all aspects of knowledge based image analysis like detection, localisation, reconstruction and identification.The paper will focus on use of GIS databases to support image analysis and the opposite, as well as fusion of multiple cues from cooperative algorithms for object extraction, reconstruction and classification.Conceptual aspects behind new developments will also be described.In general, processes exploiting different information sources often have a lower algorithmic complexity compared with single sensor data processing.This, for example, was shown with building reconstruction based on range data and a given ground plan of the buildings.Another example is map update using spectral and spatial resolution satellite images.Given a topographic map supervised classification can be executed with the result that inconsistencies between map information and image information regarding geometry and semantics can be detected and localised.With this review we aim at summarising work of our InterCommission Working Group IV/III.2 having in mind to promote further activities in this exciting field.