Vision-based UAV (Unmanned Aerial Vehicle) landing on runways is an emerging topic, because of the possible wide range of applications. Vision-based landing technology can also give an additional independent source of information for civil aviation landing systems. The core of the problem is the detection of the runway and other possible features. This paper introduces a simple framework for runway detection, in the case when only the threshold marker is painted. The approach utilizes corner features and convolutional neural nets for ROI classification.
Runway detection for UAV landing system
CNNA 2018 - The 16th International Workshop on Cellular Nanoscale Networks and their Applications ; 2018 ; Budapest, Hungary
CNNA 2018 ; 4
2018-01-01
4 pages
Conference paper
Electronic Resource
English