作者：魏永超1, 邓毅2
作者单位：1. 中国民用航空飞行学院科研处，四川 广汉 618307;
2. 中国民用航空飞行学院民航安全工程学院，四川 广汉 618307
关键词：三维点云;边界增强;关键点e-iss;k-4pcs配准
摘要：
针对点云配准出现的误匹配、耗时长、精度不高等问题，提出一种基于关键点e-iss（edge based intrinsic shape signatures）的边界增强点云配准算法。首先对源点云和目标点云进行采样、主方向变换、关键点提取、网格边界点云提取，并将特征点和边界点云融合为一个新的特征点集e-iss。对其建立快速直方图统计特征，利用算法进行配准。最后与几何采样后的配准结果、提取关键点的配准结果对比表明，所提出的改进方法在相同配准效果下，配准时间稳定在4.83 s，欧拉适应度得分为0.29，配准的均方根误差在0.093 mm内，解决了配准发动机叶片的过程中出现局部最优而导致配准失败的现象，为后续的叶片损伤定位奠定了基础。
improved point cloud alignment algorithm with key point based boundary enhancement
wei yongchao1, deng yi2
1. scientific research office, civil aviation flight university of china, guanghan 618307, china;
2. school of civil aviation safety engineering, civil aviation flight university of china, guanghan 618307, china
abstract: a boundary-enhanced point cloud alignment algorithm based on key point e-iss (edge based intrinsic shape signatures) is proposed to address the problems of mis-matching, time-consuming and low accuracy in point cloud alignment. firstly, the source and target point clouds are sampled, principal direction transformed, key points extracted, and grid boundary point clouds extracted, and the feature points and boundary point clouds are fused into a new feature point set e-iss, for which fast point feature histograms are established, and the alignment is performed using algorithm. finally, the comparison with the alignment results after geometric sampling and extraction of key points shows that the proposed improved method has a stable alignment time of 4.83 s, the euler adaptation score of 0.29, and the alignment error within 0.093 mm under the same alignment effect, which greatly solves the phenomenon of alignment failure due to local optimum in the process of aligning engine blades and provides a basis for subsequent blade damage positioning.
keywords: 3d point cloud;boundary enhancement;key point e-iss;k-4pcs registration
2023, 49(12):29-34 收稿日期: 2022-05-25;收到修改稿日期: 2022-07-22
基金项目: 国家自然科学基金（u1633127）；西藏科技厅重点研发计划（xz202101zy0017g）；中国民航飞行学院科技基金（cj2020-01）；大学生创新创业训练计划项目（s202110624114）
作者简介: 魏永超(1981-),男,四川广汉市人,教授,博士,研究方向为光电信息处理、机器视觉。
参考文献
[1] 李建微, 占家旺. 三维点云配准方法研究进展[j]. 中国图象图形学报, 2022, 27(2):349-367.
[2] 荆路, 武斌, 李先帅. 基于sac-ia和ndt融合的点云配准方法[j]. 大地测量与地球动力学, 2021, 41(4):378-381.
[3] sun j, zhang r, du s, et al. global adaptive 4-points congruent sets registration for 3d indoor scenes with robust estimation[j]. ieee access, 2020, 8:7539-7548.
[4] 张晗, 康国华, 张琪, 等. 基于改进sac-ia算法的激光点云粗配准[j]. 航天控制, 2019, 37(5):67-74.
[5] 霍旺. 基于特征点匹配的点云粗配准算法研究[d]. 太原:中北大学, 2018.
[6] theiler p w, wegner j d, schindler k. keypoint-based 4-points congruent sets-automated marker-less registration of laser scans[j]. isprs journal of photogrammetry and remote sensing, 2014, 96:149-163.
[7] huang j, kwok t h, zhou c. v4pcs:volumetric 4pcs algorithm for global registration[j]. journal of mechanical design, 2017, 139(11):111403.
[8] xu y, boerner r, yao w, et al. pairwise coarse registration of point clouds in urban scenes using voxel-based 4-planes congruent sets[j]. isprs journal of photogrammetry and remote sensing, 2019, 151:106-123.
[9] yang j, wang c, luo w, et al. research on point cloud registering method of tunneling roadway based on 3d ndt-icp algorithm[j]. sensors, 2021, 21(13):4448.
[10] pavlov a l, ovchinnikov g w v, derbyshev d y, et al. aa-icp:iterative closest point with anderson acceleration[c]//2018 ieee international conference on robotics and automation (icra). ieee, 2018:3407-3412.
[11] xu g, pang y, bai z, et al. a fast point clouds registration algorithm for laser scanners[j]. applied sciences, 2021, 11(8):3426.
[12] 张越, 翟福琪, 蔡孙宝, 等. 基于点云数据的植物叶片特征提取及三维重建[j]. 中国测试, 2021, 47(8):6-12.
[13] boroson e r, ayanian n. 3d keypoint repeatability for heterogeneous multi-robot slam[c]//2019 international conference on robotics and automation (icra). ieee, 2019:6337-6343.
[14] 叶雅欣, 王佳盛, 吴烽云, 等. 消毒机器人目标识别定位与包围盒优化[j]. 激光与光电子学进展, 2022, 59(4):346-354.
[15] rusu r b, blodow n, beetz m. fast point feature histograms (fpfh) for 3d registration[c]//2009 ieee international conference on robotics and automation. ieee, 2009:3212-3217.
