A Real-time Sensor Guided Runway Detection Method for Forward-looking Aerial Images

With the development of avionics technology, the runway detection technique becomes increasingly important both in military and civil applications. But the existing runway detection methods are mostly intended for satellite images or downward-looking aerial photos and the algorithm complexity is relatively high, which makes it hard to apply them to airborne real-time applications. In this paper, a real-time sensor guided runway detection method is presented. First of all, with the help of sensor data and terrain data from EVS and SVS, a runway template is generated and a search region is therefore determined. Then within the search region, a lines extraction on query image is performed. And finally the matching process is applied to the query image and template to detect the runway area. Experimental results on simulation system and real aerial images show that the proposed approach has a high precision and an 87.7% less time cost than an existing state-of-art matching method.