|
Author(s): |
Changxiang Lu, College of Information Science and Technology, Nanjing University of Aeronautics and Astronautics,Jiangsu Nanjing, China，210016
Yiquan Wu, College of Information Science and Technology, Nanjing University of Aeronautics and Astronautics,Jiangsu Nanjing, China，210016
Jiaming Wu, College of Information Science and Technology, Nanjing University of Aeronautics and Astronautics,Jiangsu Nanjing, China，210016
Jingjing Gu, No.93490 uints 61 uints of PLA, Jiangxi Jiujiang China，332102 |
|
Abstract: |
Combination of visible light images and SAR (synthetic aperture radar) image features, this paper presents a new algorithm for extracting the runway based on Nonsubsampled Contourlet Transform (NSCT) of the image fusion. Firstly, the same regional airports visible and SAR images are decomposed using NSCT. Then the NSCT coefficients are fused by choosing different rules which correspond to high and low frequency bands. Finally, the fused coefficients are reconstructed to obtain fusion results. The experiment shows that the fused image can reduce the disturbance, and then improved by two thresholds canny operator extracting the runway edge, and make short-term suppression section, select the best straight line. Finally, the improved Hough transform can be more accurate detection of the runway.
|