There is growing interest in performing aerial surveillance using video cam
eras. Compared to traditional framing cameras, videos provide the capabilit
y to observe ongoing activity within a scene and to automatically control t
he camera to track the activity. However, the high data rates and relativel
y small field of view of videos present new technical challenges that must
be overcome before videos can be widely used.
In this paper we present a framework and details of the key components for
real-time, automatic exploitation of aerial video for surveillance applicat
ions. The framework involves separating an aerial video into the natural co
mponents corresponding to the scene. Three major components of the scene ar
e the static background geometry, moving objects, and appearance of the sta
tic and dynamic components, of the scene. In order to delineate videos into
these scene components, we have developed real time, image-processing tech
niques for 2-D/3-D frame-to-frame alignment, change detection, camera contr
ol, and tracking of independently moving objects in cluttered scenes. The g
eo-location of video and tracked objects is, estimated by registration of t
he video to controlled reference imagery, elevation maps, and site models.
Finally static, dynamic and reprojected mosaics may be constructed for comp
ression, enhanced visualization, and mapping applications.