The work aims to propose a 3D point cloud feature extraction method combining multi-view depth map projection and sub-pixel edge detection, and support the structural optimization design and safety verification of containers with high-precision geometric feature data, so as to solve the bottleneck problem of insufficient accuracy of 3D deformation measurement in reliability evaluation of radioactive materials transport containers. Firstly, by constructing the minimum bounding cube and designing a six-degree-of-freedom multi-view orthogonal projection system, the three-dimensional point cloud was mapped to a two-dimensional depth map to realize the full surface coverage of the container. Secondly, based on the fitting edge model and local gray area effect, sub-pixel edge points were detected; Furthermore, combining projection parametric mapping with dynamic curvature interpolation algorithm, normal vector continuity constraint and weighted fusion strategy were introduced to reconstruct high-fidelity three-dimensional feature points; Finally, an experimental system was built to obtain the 3D point cloud data of the container, and compared with the existing point cloud feature extraction methods based on simplification rate, registration error and calculation time. The experimental results showed that the maximum average error, the maximum root mean square error and the simplification rate of the feature point cloud obtained by the proposed method were 1.33 mm, 1.39 mm and 1.6%, respectively. Compared with other literature results, the proposed method had a smaller simplification rate, faster calculation time and smaller error. The method proposed in this study effectively solves the contradiction between large-scale point cloud processing and geometric feature fidelity, significantly improves the accuracy and efficiency of three-dimensional feature extraction of radioactive containers, and provides reliable technical support for geometric deformation quantitative analysis for nuclear energy engineering safety assessment.
Key words
transport containers for radioactive materials /
three-dimensional deformation measurement /
multi-view depth map fusion /
sub-pixel edge detection
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
References
[1] 孙谦, 庄大杰, 孙洪超, 等. 核燃料组件运输容器应用现状概述[J]. 包装工程, 2022, 43(13): 142-150.
SUN Q, ZHUANG D J, SUN H C, et al.Overview of Application Status for Nuclear Fuel Assembly Transport Package[J]. Packaging Engineering, 2022, 43(13): 142-150.
[2] KIM K S, KIM J S, CHOI K S, et al.Dynamic Impact Characteristics of KN-18 SNF Transport Cask - Part 1: An Advanced Numerical Simulation and Validation Technique[J]. Annals of Nuclear Energy, 2010, 37(4): 546-559.
[3] BARTOL K, BOJANIĆ D, PETKOVIĆ T, et al.A Review of Body Measurement Using 3D Scanning[J]. IEEE Access, 2021, 9: 67281-67301.
[4] YAO Z W, ZHAO Q X, LI X F, et al.Point Cloud Registration Algorithm Based on Curvature Feature Similarity[J]. Measurement, 2021, 177: 109274.
[5] WANG G L, WU L S, HU Y, et al.Point Cloud Simplification Algorithm Based on the Feature of Adaptive Curvature Entropy[J]. Measurement Science and Technology, 2021, 32(6): 065004.
[6] XU Y S, TONG X H, STILLA U.Voxel-Based Representation of 3D Point Clouds: Methods, Applications, and Its Potential Use in the Construction Industry[J]. Automation in Construction, 2021, 126: 103675.
[7] XIE Z P, LANG Y C, CHEN L Q.Geometric Modeling of Rosa Roxburghii Fruit Based on Three-Dimensional Point Cloud Reconstruction[J]. Journal of Food Quality, 2021, 2021(1): 9990499.
[8] SUN Y, ZHANG S H, WANG T Q, et al.An Improved Spatial Point Cloud Simplification Algorithm[J]. Neural Computing and Applications, 2022, 34(15): 12345-12359.
[9] LEAL E, SANCHEZ-TORRES G, BRANCH-BEDOYA J W, et al. A Saliency-Based Sparse Representation Method for Point Cloud Simplification[J]. Sensors, 2021, 21(13): 4279.
[10] GUMHOLD S, WANG X, MACLEOD R S.Feature Extraction from Point Clouds[C]//IMR, 2001: 293-305.
[11] HO H T, GIBBINS D.Curvature-Based Approach for Multi-Scale Feature Extraction from 3D Meshes and Unstructured Point Clouds[J]. IET Computer Vision, 2009, 3(4): 201-212.
[12] WANG L, YUAN B.Feature Point Detection for 3D Scattered Point Cloud Model[J]. Signal Process, 2011, 27(6): 932-938.
[13] WANG S Q, HU Q J, XIAO D S, et al.A New Point Cloud Simplification Method with Feature and Integrity Preservation by Partition Strategy[J]. Measurement, 2022, 197: 111173.
[14] POTAMIAS R A, BOURITSAS G, ZAFEIRIOU S.Revisiting Point Cloud Simplification: A Learnable Feature Preserving Approach[M]//Computer Vision - ECCV 2022. Cham: Springer Nature Switzerland, 2022: 586-603.
[15] WU J S, LAI X M, CHAI X L, et al.Feature-Based Point Cloud Simplification Method: An Effective Solution for Balancing Accuracy and Efficiency[J]. The Journal of Supercomputing, 2024, 80(10): 14120-14142.
[16] ALSHAWABKEH Y.Linear Feature Extraction from Point Cloud Using Color Information[J]. Heritage Science, 2020, 8(1): 28.
[17] HU H L, LI Z, JIN X G, et al.Curve Skeleton Extraction from 3D Point Clouds through Hybrid Feature Point Shifting and Clustering[J]. Computer Graphics Forum, 2020, 39(6): 111-132.
[18] SUN Y, ZHANG S H, WANG T Q, et al.An Improved Spatial Point Cloud Simplification Algorithm[J]. Neural Computing and Applications, 2022, 34(15): 12345-12359.
[19] DING Z H, ZHENG S H, ZHANG F L, et al.Lensfree Auto-Focusing Imaging with Coarse-to-Fine Tuning Method[J]. Optics and Lasers in Engineering, 2024, 181: 108366.
[20] TRUJILLO-PINO A, KRISSIAN K, ALEMÁN-FLORES M, et al. Accurate Subpixel Edge Location Based on Partial Area Effect[J]. Image and Vision Computing, 2013, 31(1): 72-90.
[21] DU H, HU Z Y, LU D P, et al.Density Clustering Algorithm Based on KD-Tree and Voting Rules[J]. Computers, Materials & Continua, 2024, 79(2): 3239-3259.
[22] QI J K, HU W, GUO Z M.Feature Preserving and Uniformity-Controllable Point Cloud Simplification on Graph[C]//2019 IEEE International Conference on Multimedia and Expo (ICME). IEEE, 2019: 284-289.
PDF(10400 KB)
