Semi-automatic verification of geodata for quality management and updating of gis

The usefulness and acceptance of geoinformation systems mainly depend on the quality of the underlying geodata. The aim of this paper is to introduce a system for the verification of geodata. The verification of geodata can be seen as part of quality management, and furthermore, of the update of geodata systems (GIS). The presented system compares GIS data with reference data derived from, e.g. high resolution remote sensing imagery using image analysis operators. After the detection of evidence an object will be accepted or rejected by the system. It works semi-automatic, the final decision is made by a human operator. The verification system is presented in detail, and examples are given for various methods for feature extraction. In addition, we illustrate the potential of the system, but we show the limits of this system, too.