Site Model Based Image Registration and Change Detection - RADIUS

Abstract : The University of Maryland (with TASC as a subcontractor) is one of a group of institutions doing research on aerial image understanding in support of the RADIUS program. The emphasis of our research is on knowledge-based change detection (CD) using site models and the domain expertise of image analysts IAs) . We are designing a system that allows the IA to specify what are to be considered as significant changes through quick look (QL) profiles, and to select appropriate image understanding algorithms for detecting these changes. Before CD can be attempted, the acquired images have to be registered to the site model. Two algorithms for image registration have been developed. When no information about the camera is available, we use an efficient constrained search mechanism for image-to-image registration. When an approximate camera model is available, as in RADIUS applications, we use a fast image-to-site model registration algorithm which first projects the site model into the new image domain using the given approximate camera model, and then use five control points to do camera resection and to obtain an accurate camera mode. To enable efficient transfer of technology to the lAs, we have developed our algorithms under the RADIUS Common Development Environment (RCDE). Aerial image understanding, Change detection, Image analysis, Quick-look profile, Registration