Abstract:
Skyline detection plays an important role in unmanned aerial vehicle visual flight control systems. Many studies assume that the skyline in the image is a straight line w...Show MoreMetadata
Abstract:
Skyline detection plays an important role in unmanned aerial vehicle visual flight control systems. Many studies assume that the skyline in the image is a straight line with clear edge features, but this assumption is not necessarily true under different weather and environmental conditions. For example, when it is cloudy or foggy the contrast and brightness of the image becomes weaker and the skyline is not obvious. The edge features of skyline are hard to be found in these cases. When an aircraft is flying at low altitudes, mountains, woods, or buildings become part of the skyline. Because of this, the contour of the skyline is not a straight line anymore, and the assumptions contradict the actual situation. In order to solve these problems, this paper proposes an algorithm that can detect the contour of a curving skyline in an image under different brightnesses and contrasts. Experiments are conducted using 40 images taken in different weather and environmental conditions. The proposed algorithm can achieve stable skyline results in different weather and environmental conditions. The results can be used as a reference for unmanned aerial vehicle flight in order to determine flight attitude.
Date of Conference: 19-21 May 2016
Date Added to IEEE Xplore: 08 August 2016
ISBN Information:
Electronic ISSN: 2154-0373