A multiresolution modelling approach for semi-automatic extraction of streets: application to high resolution images from the IKONOS satellite
Résumé
In order to help people working in the urban areas, we are concerned about extracting road networks in from high resolution satellite images. However, due to the urban landscape complexity, the development of reliable and robust algorithms to extract information from city images is difficult. An existing street extraction method is applied on hi high-resolution images from the IKONOS satellite. Results of this method are discussed. Then, a methodology is proposed which aims at semi-automatically extracting all types of urban street networks and at managing their topology.