Surface Defect Detection on Nuclear Power Plant Components Based on Photometric Stereo under Near-Field LED Light
-
摘要: 为提高核电设备表面缺陷检测能力，研究了近场发光二极管（LED）点光源照明条件下的光度立体三维检测方法。该方法采用迭代算法，确定光源发光特性参数，进而实现精确的光照强度估计，并结合光源与被检表面点空间位置的计算方法，实现近场LED点光源照明下被检测面上不同点的光照强度与光线方向的估计。以此为基础设计表面缺陷三维检测系统，并将该系统在表面损伤试样以及实际核电设备上进行实验验证。结果表明，该系统可以获取表面缺陷三维信息，并且对于划伤类缺陷，能够实现比较精确的深度测量。因此，该系统可以有效提高表面缺陷的检测能力。
-
关键词:
- 光度立体 /
- 近场发光二极管（LED）点光源 /
- 光照强度估计 /
- 核电设备 /
- 表面缺陷
Abstract: To improve the visual inspection of nuclear power plant components, this study introduces the technology of three-dimensional reconstruction framework based on photo metric stereo under near-field LED lighting. An iterative algorithm is proposed to estimate the K value of different light sources and the light6 intensities for various image pixels. Combined with the calculation method of the spatial position of the light source and the detected surface points, the light intensity and light direction of different points on the detected surface are estimated under the illumination of near-field LED point light source. A surface defect detection system is designed based on this technology, and it is verified by experiments on surface damage samples and actual nuclear power equipment. The results show that this system can obtain three-dimensional surface data, and for scratch-type defects, it can further achieve accurate depth measurement. Therefore, it can effectively improve the detection ability of surface defects. -
图 1 近场光度立体相机光源布置示意图
OcXcYcZc—相机坐标系；OXwYwZw—世界坐标系；L1、L2、L3—表面某一点对应的其中3个光源的光线方向
Figure 1. Diagram of Photometric Stereo System
图 2 表面损伤试样
编号1、2、3—深度为1、2、3 mm的3条划伤；编号4、5—不同直径与深度的2个压痕
Figure 2. Specimen of Surface Damage
图 3 1号划伤法向量图与三维重建结果
x—沿缺陷方向的长度；y—垂直缺陷方向的宽度；z—垂直缺陷平面的深度；1 pixel=0.077 mm
Figure 3. Normal and 3D Reconstruction Result of Defect No. 1
图 8 压力容器管嘴安全端加工件表面缺陷
①、②、③、④—缺陷编号
Figure 8. Surface Defects of Machined Parts of Safety End of Pressure Vessel Nozzle
图 9 安全端加工件表面缺陷三维形貌重建
Figure 9. 3D Reconstruction Results of Surface Defects of Machined Parts of Safety End
表 1 划伤定量测量 单位：mm
Table 1. Size of Scratches
样本编号 深度测量 宽度测量 深度误差 宽度误差 1 1.27 1.26 0.27 0.26 2 1.86 2.05 −0.14 0.05 3 2.86 3.12 −0.14 0.12 
下载: 导出CSV
-
[1] KANG D Y, JANG Y J, WON S C. Development of an inspection system for planar steel surface using multispectral photometric stereo[J]. Opt. Eng., 2013(52): 039701-039701. doi: 10.1117/1.OE.52.3.039701 [2] 徐科,周鹏,杨朝霖. 基于光度立体学的金属板带表面微小缺陷在线检测方法[J]. 机械工程学报,2013, 49(4): 25-29. [3] EVA W, SEBASTIAN Z, MATTHIAS S, et al. Photometric stereo sensor for robot-assisted industrial quality inspection of coated composite material surfaces[C]. In Proceedings of the SPIE 12th International Conference on Quality Control by Artificial Vision, Le Creusot, France, 2015. [4] 王磊,徐科,周鹏,等. 利用多尺度小波变换的光度立体快速三维表面重建算法[J]. 计算机辅助设计与图形学学报,2017, 29(1): 124-129. doi: 10.3969/j.issn.1003-9775.2017.01.015 [5] WOODHAM R J. Photometric method for determining surface orientation from multiple images[J]. Optical Engineering, 1980, 19(1): 1-22. [6] XIE L M, SONG Z H, JIAO G H, et al. A Practical means for calibrating an LED-based photometric stereo system[J]. Opt. Lasers Eng., 2015(64): 42-50. [7] FAN H, QI L, WANG N, et al. Deviation correction method for close-range photometric stereo with nonuniform illumination[J]. Optical Engineering, 2017, 56(10): 103102. [8] 谢利民. 基于近场照明的光度立体视觉算法研究[D]. 武汉: 华中科技大学, 2015. [9] HUANG S A, XU K, LI M, et al. Improved visual inspection through 3D image reconstruction of defects based on the photometric stereo technique[J]. Sensors., 2019, 19(22): 4970. doi: 10.3390/s19224970 [10] QUEAU Y, DURIX B, WU T, et al. LED-based photometric stereo: modeling, calibration and numerical solution[J]. Math. Imaging Vis., 2018(60): 1-26. [11] LOGOTHETIS F, MECCA R, CIPOLLA R. Semi-calibrated near field photometric stereo[C]. In Proceedings of the CVPR 2017, Honolulu, HI, USA, 2017. [12] QUEAU Y, WU T, CREMERS D. Semi-calibrated near-light photometric stereo[C]. In Proceedings of the SSVM 2017, Kolding, Denmark, 2017. [13] HAN T Q, SHEN H L. Photometric stereo for general BRDFs via reflection sparsity modeling[J]. IEEE Trans. Image Process., 2015(24): 4888-4903. [14] 李敏. 基于非朗伯光度立体的三维重建研究[D]. 杭州: 浙江大学, 2020. [15] SARACCHINI R F V, STOLFI J, LEITAO H C G, et al. A robust multi-scale integration method to obtain the depth from gradient maps[J]. Computer Vision and Image Under-standing, 2012, 116(8): 882-895. doi: 10.1016/j.cviu.2012.03.006 -
计量
- 文章访问数: 619
- HTML全文浏览量: 253
- PDF下载量: 51
- 被引次数: 0
