A Strategy of Change Detection Based on Remotely Sensed Imagery and Gis Data

Geometric co-registration errors and results of object extraction (such as classification and pattern recognition) are the two main factors, which greatly affect the results of change detection in traditional methods. Unfortunately, the geometric co-registration accuracy and results obtained by the object extraction methods can not satisfy imagery change detection in practice at present. Considering the change detection based on remotely sensed imagery and GIS data, GIS data can direct detection, because our interesting objects are included in the GIS data. To overcome the influences of geometric co-registration and object extraction errors, a new change detection method is needed to develop. The paper proposes a holistic solution strategy, which is based on areal features and iteratively solves geometric co-registration, feature extraction and change detection simultaneously. Some concepts and key techniques are introduced in the paper. Finally, the data of a QUICKBIRD image and corresponding 1:2000 topographic map of a district in Shanghai, China is chosen to test the strategy presented in the paper. The change detection precision of 83.33% is gotten, that proves the new strategy is efficient and applicable.