Research on Aircraft Skin Defect Identification Method based on Point Cloud Processing

Baihui Li ； Yongxiang Jiang ； Hongchang Sun ； Yingwei Qiao ； Jinchang Guo

《International Core Journal of Engineering》 10卷1期 (2024/01) Pp. 345-349

https://doi.org/10.6919/ICJE.202401_10(1).0048

摘要

The outer surface of aircraft skin may be subjected to external impacts during aircraft service, such as hail, stones, birds, etc., which can cause surface depressions and have very serious consequences. In order to detect these defects, this article uses a binocular camera to obtain damaged skin point clouds and detect defects. Firstly, the binocular camera was calibrated using Zhang Zhengyou's calibration method. Then, the defect point cloud was preprocessed by point cloud filtering and point cloud simplification, and a standard point cloud model was fitted using the defective point cloud; The point cloud feature descriptor sub algorithm based on Fast PointFeature Histogram (FPFH) and the Nearest Point Iteration (ICP) algorithm were used to perform coarse and fine registration on two point clouds. Finally, the Euclidean distance was calculated on the registered point clouds to complete defect detection.

關鍵字

Binocular Vision ； Point Cloud Denoising ； Point Cloud Registration ； Defect Recognition

參考文獻

Zong Yulong, Liang Jin, Wang Huan, et al. An intelligent and automated 3D surface defect ection system for quantitative 3D estimation and feature classification of material surface ects[J]. Optics and Lasers in Engineering, 2021, 144: 106633.

Zhao Y S, Liu J, Wang G F, et al. 3D concave defect measurement system of the cryogenic insulated cylinder based on linear structured light[C]//2018 Chinese Control and Decision Conference (CCDC). 2018.

Li Z X, Fang X, Shao X L, et al. Rotor surface defect and symmetry detection based on point cloud registration[C]//2021 IEEE International Conference on Power Electronics, Computer Applications (ICPECA). IEEE, 2021.

Zhang Bin, Xiong Chuanbing, Automatic Point Cloud Registration Based on Voxel Downsampling and Key Point Extraction[J]. Laser & Optoelectronics Progress, 2020, 57(4): 109-117.

Liu Meiiu, Wang Xudong, Li Lingyan, et al. lmproved Random Sampling Consistency Algorithm Employed in Three-Dimensional Point Cloud Registration[J]. Laser & Optoelectronics Progress, 2018, 55(10):165-171.