Most vision based UAV (Unmanned Aerial Vehicle)
navigation algorithms extract features such as horizons
and mountain peaks from 2D input images, and match the
extracted features with features obtained from
DEM(Digital Elevation Map) by process of registration.
The difficulties of the horizon and peak extraction originate
from the variations of input images such as noise,
viewing direction, and scale. Moreover to prove the existence
of horizon is also difficult. Therefore the success of
the feature extraction will depend on its ability to cope
with these variations. In this paper, we present a new feature
extraction method, which is robust to these variations
and verified throughout the following experiments.