Automatic registration of visible-band and SAR satellite and aerial remotely sensed images.

摘要

Image registration is an essential and difficult task where dealing with remotely sensed images. There exist different registration techniques that work with different data. However, no algorithm is known that can accurately register different type sensor's images consistently. This thesis addresses this problem by investigating the development of a fully automatic registration system for synthetic aperture radar (SAR) and optical remotely sensed images. This novel automatic image registration method is based on the extraction and matching of common features that are visible in the image pairs. The algorithm involves the following five steps: noise removal, edge extraction; edge linking, object extraction, and object matching. The application of the proposed automatic image registration model to SAR and visible-band image pairs shows that accurate ground control points (GCPs) can be identified automatically.