本論文的目的在於建立一套適合應用於無人飛行載具，並且適用於不同的降落環境的跑道辨識方法。本文敘述如何將無人機由遠處導引至降落跑道周邊，進一步利用機器視覺進行跑道辨識。
由於跑道的顏色與形狀往往與周圍環境有明顯區隔，因此以顏色特徵作為研究的出發點。此外在不同光照下，不同顏色、材質的物體所顯現的特徵皆不同，由於HSV特徵較RGB色彩空間更不受陰影干擾，因此採用HSV作為辨別跑道與環境的首要方法。另一方面，以LSD作為線段偵測的工具以辨別跑道邊線，運算速度較常見的Canny邊緣偵測等等方法快，並且也有更好的偵測成果。

Current autoland system in civil aviation is complicated, which needs onboard navigation and autopilot instruments as well as ground-based navigation equipment, such as localizer (LOC) and glide slope (GS). The purpose of this thesis is to build a runway detecting method which is adequate to different landing environment for UAVs. The approach of guiding UAVs to landing area and then detecting runway will be mentioned in this thesis.
This thesis begins with color feature because the color and shape feature of a runway are usually different from the background. In addition, objects of different colors and materials will have different characteristic under different light sources. Also the Hue-Saturation-Value (HSV) color space will give more precise result than Red, Green, and Blue (RGB) color space when in under shaded condition. Therefore the HSV is used as the first step for separating runway and background. On the other hand, using Line Segment Detector (LSD) for runway edge detection has better detecting result and has better computational efficiency than using Canny edge detector.

[1] C2Land @ TU Braunschweig, available in April 2015 from website: https://www.tu-braunschweig.de/iff/forschung/c2land
[2] J. J. Shang and Z. K. Shi, “Vision-based Runway Recognition for UAV Autonomous Landing,” International Journal of Computer Science and Network Security, Vol. 7, No. 3, March 2007, pp. 112-117.
[3] P. Y. Huang, “Realization of Vision-based Runway Detection for Fixed-wing UAV,” Master's Thesis, Institute of Aeronautics and Astronautics, National Cheng Kung University, 2009.
[4] G. S. Rajput, Z. Rahman, “Hazard Detection on Runways Using Image Processing Techniques,” Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 6957, April 2008.
[5] Cascade Classifier Training - opencv documentation, available in April 2015 from website: http://docs.opencv.org/doc/user_guide/ug_traincascade.html
[6] Gaussian blur - Wikipedia, the free encyclopedia, available in April 2015 from website: http://en.wikipedia.org/wiki/Gaussian_blur
[7] P. A. Viola and M. J. Jones, “Rapid Object Detection using a Boosted Cascade of Simple Features,” IEEE Conference on Computer Vision and Pattern Recognition, Vol. 1, 2001.
[8] R. Maini and H. Aggarwal, “Study and Comparison of Various Image Edge Detection Techniques,” International Journal of Image Processing (IJIP), Vol. 3, 2009, pp. 1-11.
[9] R. G. Gioi, J. Jakubowicz, J.-M. Morel, and G. Randall, “LSD: A Fast Line Segment Detector with a False Detection Control,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol 32, 2010, pp. 722-732.
[10] R. G. Gioi, J. Jakubowicz, J.-M. Morel, and G. Randall, “LSD: A Line Segment Detector,” Image Processing On Line, Vol 2, 2012, pp. 35–55.
[11] S. K. Yang, “Genetic Algorithm Path Planning with Camera Gimbal Tracking for UAV Flight Control,” Master’s Thesis, Institute of Aeronautics and Astronautics, National Cheng Kung University, Taiwan, 2014.
[12] Y. R. Huang, Y. L. Pan, G. Z. Wu, “Research on Road Surface Detection without Lane Markings,” International Conference on Advanced Information Technologies (AIT), April 2009.