Détection automatique des changements du bâti en milieu urbain sur des images à très haute résolution spatiale (Ikonos et QuickBird) en utilisant des données cartographiques numériques
The updating of cartographic databases in urban environments is a difficult and expensive task. It can be facilitated by an automatic change detection method. Several methods have been developed for medium and low spatial resolution images. These methods are not adapted for the very high spatial resolution images (VHSR) and are not applicable in urban environment. This study proposes a new method for change detection of buildings in urban environments from VHSR images and using existing digital cartographic data. The proposed methodology is composed of several stages. The existing knowledge on the buildings and the other urban objects are first modelled and saved in a knowledge base. All change detection rules are defined at this stage. Then, the image is segmented. The parameters of segmentation are computed thanks to the integration between the image and the geographical database (GDB). Thereafter, the segmented image is analyzed using the knowledge base to localize the segments where the change of building is likely to occur. The change detection rules are then applied on these segments to identify the segments that represent the changes of buildings. These changes represent the updates of buildings to add to the geographical database. Finally, the map representing changes is assessed before being integrated in the geographical database. The data used in this research concern the city of Sherbrooke (Quebec, Canada) and the city of Rabat (Morocco). For Sherbrooke, we used an Ikonos image acquired in October 2004, an Ikonos image acquired in July 2006 and a GDB at the scale of 1:20,000. For Rabat, a QuickBird image acquired in August 2004 has been used with a GDB at the scale of 1:10,000. The results of tests on several zones are encouraging. Indeed, the rate of good detection is of 90%. Concerning the geometric precision of detection, the mean error is 3 m for Ikonos and 2 m for QuickBird. The proposed method presents some limitations on the detection of the exact contours of the buildings. It could be improved by including a shape post-analysis of detected buildings. The proposed method can be integrated in a cartographic update process or as a method for the quality assessment of a topographic database.