Aerodrome Taxiway Line Detection and Cross-Track Error Estimation using Computer Vision Techniques

Abstract

This paper describes a computer vision-based method to detect aerodrome taxiway lines and to estimate the deviation of a large passenger aircraft from the taxiway centerline using an onboard camera located on the vertical stabilizer of the aircraft. This method could be applied as part of a larger system to increase the situation awareness of pilots during taxiing and to alert them if the aircraft deviates from the centerline. The proposed method takes advantage of color and edge information in the camera images and proposes a Sliding Window (SW) method and clustering techniques to detect and process taxiways markings. First, the input image is transformed to a top-down view by applying the Homographic Transform. Then, color and edge detection techniques are applied to the top-down view to generate a binary map of pixels belonging to taxiway lines. Then, the region of the image directly in front of the aircraft is processed to determine whether the aircraft is turning or moving in a straight line. If it is determined that the aircraft is turning, the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering technique is applied to the binary map; otherwise, a SW method is used to detect the taxiway centerline ahead of the aircraft and to detect any taxiway branches or junctions. Finally, the aircraft's lateral deviation from the taxiway centerline is estimated using a template matching approach. Tests on simulated video sequences show that the SW technique achieves a detection rate of 80% and a false positive rate of 3%. On the other hand, the clustering technique achieves a detection rate of 76% and a false positive rate of 4%. The aircraft's deviation from the taxiway centerline is estimated with a mean error of 0.8 pixels and a worst case error of ±3 pixels.