Title :
Research on Aircraft Skin Cracks Recognition Based on the Crack-Isolated Method
Author :
Gao, Qing-Ji ; Li, Li ; Niu, Guo-Chen
Author_Institution :
Dept. of Aeronaut. Autom., Civil Aviation Univ. of China, Tianjin, China
Abstract :
A crack-isolated method is proposed for the detection of aircraft skin crack defects. It takes the rivet detection based on the magneto-optic images as the clue to detect skin cracks. Firstly, the rivet sites are classified according to the circularity of candidate sites after images pre-processing. Secondly, a circle template is constructed by the rivet center and radius to cover rivet sites, and then the potential crack regions are isolated from the template region. Finally, cracks´ existence and direction are determined according to isolated regions´ area and shape. The experimental results prove the effectiveness of the crack-isolated method in segmenting crack regions and detecting multi-cracks.
Keywords :
aircraft; crack detection; image recognition; image segmentation; aircraft skin crack defect detection; aircraft skin crack recognition; circle template; crack region segmentation; crack-isolated method; image preprocessing; magneto-optic images; rivet detection; Additive noise; Aging; Aircraft; Automation; Image recognition; Noise generators; Probes; Sensor phenomena and characterization; Shape; Skin;
Conference_Titel :
Biomedical Engineering and Computer Science (ICBECS), 2010 International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-5315-3
DOI :
10.1109/ICBECS.2010.5462507
