|
|
- 2018
直线特征约束下利用Plücker坐标描述的LiDAR点云无初值配准方法
|
Abstract:
经典的基于点状特征匹配的地面激光雷达（light detection and ranging，LiDAR）点云配准算法实现过程中，点状特征的提取精度对算法运行结果的影响通常较大；基于迭代运算的LiDAR点云配准算法计算量大，对未知参数的初值依赖程度较高，在求解大转角刚体变换参数时算法不稳定。对此，提出了一种线状特征约束下基于Plücker直线坐标描述的LiDAR点云配准算法。立足于经典的向量代数与对偶四元数的相关理论与方法，分析并确定了Plücker直线坐标与对偶四元数之间的相互转换关系以及模型描述方法；以LiDAR点云配准前后同名线状特征的Plücker直线坐标相等为约束条件，构建了线状特征约束下基于Plücker直线坐标描述的刚体变换模型；立足于最小二乘基本准则，通过目标函数的极值化分析实现了线状特征约束下地面LiDAR点云配准参数的直接求解。实验结果表明，所构建的基于Plücker直线坐标描述的地面LiDAR点云配准模型，无需事先确定变换参数的初值，避免了多元函数的线性化过程，解除了参数结果对于迭代初值的依赖，理论上克服了迭代法在求解大转角相似变换参数时的算法不稳定问题。此外，较之单纯基于点状特征匹配的LiDAR点云配准算法，该算法可以有效地增强LiDAR点云配准过程的约束，达到提高配准质量的目的
| [1] | Horn B K. Closed Form Solution of Absolute Orientation Using Orthonormal Matrices[J]. <em>Journal of Optical Society of America, Series A</em>, 1988, 5(7):1127-1135 |
| [2] | Arun K S, Huang T S. Least-Squares Fitting of Two 3D Point Sets[J]. <em>IEEE Trans Pattern Analysis and Machine Intelligence</em>, 1987, 9(5):698-700 |
| [3] | Horn B K. Closed-Form Solution of Absolute Orientation Using Unit Quaternions[J]. <em>Journal of Optical Society of America, Series A</em>, 1987, 4(4):629-642 |
| [4] | Pro？ková J. Discovery of Dual Quaternions for Geo-desy[J]. <em>Journal for Geometry and Graphics</em>, 2012, 16(2):195-209 |
| [5] | He F, Habib A. A Closed-Form Solution for Coarse Registration of Point Clouds Using Linear Features[J]. <em>Journal of Survey and Engineering</em>, 2016, 142(3):04016006,DOI:10.1061/(ASCE) SU.1943-5428. 0000174 |
| [6] | Sheng Qinghong, Chen Shuwen, Liu Jianfeng, et al. LiDAR Point Cloud Registration Based on Plücker Line[J]. <em>Acta Geodaetica et Cartographica Sinica</em>, 2016, 45(1):58-64(盛庆红, 陈姝文, 柳建锋, 等.基于Plücker直线的LiDAR点云配准法[J].测绘学报, 2016, 45(1):58-64) |
| [7] | Khoshelham K. Closed-Form Solutions for Estimating a Rigid Motion from Plane Correspondences Extracted from Point Clouds[J]. <em>ISPRS Journal of Photogrammetry and Remote Sensing</em>, 2016, 114:78-91 |
| [8] | Zheng Dehua, Yue Dongjie, Yue Jianping. Geome-tric Feature Constraint Based Algorithm for Buil-ding Scanning Point Cloud Registration[J]. <em>Acta Geodaetica et Cartographica Sinica</em>, 2008, 37(4):464-468(郑德华, 岳东杰, 岳建平. 基于几何特征约束的建筑物点云配准算法[J]. 测绘学报, 2008, 37(4):464-468) |
| [9] | Shen Y Z, Chen Y, Zheng D H. A Quaternion-Based Geodetic Datum Transformation Algorithm[J]. <em>Journal of Geodesy</em>, 2006, 80:233-239 |
| [10] | Pro？ková J. Application of Dual Quaternions Algorithm for Geodetic Datum Transformation[J]. <em>Journal of Applied Mathematics</em>, 2011, 4(2):225-236 |
| [11] | Pottmann H, Hofer M, Odehnal B, et al. Line Geometry for 3D Shape Understanding and Reconstruction[C]. European Conference on Computer Vision, Prague, Czech Republic, 2004 |
| [12] | Walker M W, Shao L, Volz R A. Estimating 3D Location Parameters Using Dual Number Quater-nions[J]. <em>CVGIP:Image Understanding</em>, 1991, 54(3):358-367 |
| [13] | Gong Hui, Jiang Gangwu, Jiang Ting, et al. Close-Form Solution of Absolute Orientation Based on Dual Quaternion[J]. <em>Journal of Geomatics Science and Technology</em>, 2009, 26(6):434-438(龚辉, 江刚武, 姜挺, 等. 基于对偶四元数的绝对定向直接解法[J]. 测绘科学技术学报, 2009, 26(6):434-438) |
| [14] | Wang Yongbo, Wang Yunjia, Wu Kan, et al. A Dual Quaternion-Based, Closed-Form Pairwise Re-gistration Algorithm for Point Clouds[J]. <em>ISPRS Journal of Photogrammetry and Remote Sensing</em>, 2014, 94:63-69 |
| [15] | Habib A, Mwafag G, Michel M, et al. Photogrammetric and LiDAR Data Registration Using Linear Features[J]. <em>Photogrammetric Engineering & Remote Sensing</em>, 2005, 71(6):699-707 |
| [16] | Daniilidis K. Hand-Eye Calibration Using Dual Quaternions[J]. <em>The International Journal of Robotics Research</em>, 1999, 18(3):286-298 |
| [17] | Wang Yongbo, Yang Huachao, Liu Yanhua, et al. Linear-Feature-Constrained Registration of LiDAR Point Cloud via Quaternion[J]. <em>Geomatics and Information Science of Wuhan University</em>, 2013, 38(9):1057-1062(王永波, 杨化超, 刘燕华, 等. 线状特征约束下基于四元数描述的LiDAR点云配准方法[J]. 武汉大学学报·信息科学版, 2013, 38(9):1057-1062) |
| [18] | Renaudin E, Habib A, Kersting A. Featured-Based Registration of Terrestrial Laser Scans with Minimum Overlap Using Photogrammetric Data[J]. <em>ETRI Journal</em>, 2011, 33(4):517-527 |
