Feature Matching For UAV Navigation in Urban Environments

This research was motivated by a need for accurate UAV navigation using camera(s) to augment inertial navigation unit data while flying over, or through an urban environment. The process of position determination using cameras follows a sequence of well defined steps. First features must be found in consecutive-in-time images and then matched across time. The underlying camera motion, as defined by camera rotation and translation, may then be calculated as a homography, for example. Accurate homography determination requires use of a set of at least 4 feature pairs that are not collinear and are accurately matched. In this paper it is shown that collinearity of features in urban images is quite likely to occur. Thus, collinear features must be removed. This paper reports an easy way to do so and examines, as well, the issue of accurate matching. Preliminary results are reported.