Real-time Accurate Runway Detection based on Airborne Multi-sensors Fusion

Existing methods of runway detection are more focused on image processing for remote sensing images based on computer vision techniques. However, these algorithms are too complicated and time-consuming to meet the demand for real-time airborne application. This paper proposes a novel runway detection method based on airborne multi-sensors data fusion which works in a coarse-to-fine hierarchical architecture. At the coarse layer, a vision projection model from world coordinate system to image coordinate system is built by fusing airborne navigation data and forward-looking sensing images, then a runway region of interest (ROI) is extracted from a whole image by the model. Furthermore, EDLines which is a real-time line segments detector is applied to extract straight line segments from ROI at the fine layer, and fragmented line segments generated by EDLines are linked into two long runway lines. Finally, some unique runway features (e.g. vanishing point and runway direction) are used to recognise airport runway. The proposed method is tested on an image dataset provided by a flight simulation system. The experimental results show that the method has advantages in terms of speed, recognition rate and false alarm rate.