高维正交子空间映射的尺度点云配准算法
|
摘要
为了解决三维点云在无序、数据被遮挡以及噪声干扰情况下的配准问题,提出了一种高维正交子空间映射的尺度点云配准算法。根据能量-功率的比值,对待配准点云进行等比例放大,完成仿射配准。在点云无序、数据被遮挡、尺寸放缩以及噪声干扰的情况下,所提算法与经典ICP(Iterative Closest Point)算法的配准精度相当。与经典ICP算法相比,所提算法对Bunny点云数据的配准效率提高了98%,对Dragon点云数据的配准速度至少提高了20倍,且在对大尺度Dragon点云数据的配准中,所提算法的配准时间比经典ICP算法短6210.4 s,配准精度也高于其他算法。所提算法不会陷入局部最小值,在快速精确配准和稳定性方面有明显的优势。
To solve the registration problem of a three-dimensional(3 D) point cloud under disorder, data occlusion and noise disturbance, a scale point cloud registration algorithm in high-dimensional orthogonal subspace mapping is proposed. The point cloud to be registered is scaled up to complete the affine registration according to the energy-power ratio. The registration accuracy of the proposed algorithm is comparable to that of the classical iterative closest point(ICP)algorithm when the point cloud is out of order with data occluded, size scaled and noise disturbance. Compared with the classical ICP algorithm, the proposed algorithm improves the registration efficiency of the Bunny point cloud data by 98% and the registration speed of the Dragon point cloud data by at least 20 times. Moreover, in the registration of the large-scale Dragon point cloud data, the registration time of the proposed algorithm is 6210.4 s less than that of the classical ICP algorithm, and the registration accuracy is higher than those of other algorithms. The proposed algorithm does not fall into the local minimum and possesses obvious advantages in terms of fast and accurate registration and stability.
To solve the registration problem of a three-dimensional(3 D) point cloud under disorder, data occlusion and noise disturbance, a scale point cloud registration algorithm in high-dimensional orthogonal subspace mapping is proposed. The point cloud to be registered is scaled up to complete the affine registration according to the energy-power ratio. The registration accuracy of the proposed algorithm is comparable to that of the classical iterative closest point(ICP)algorithm when the point cloud is out of order with data occluded, size scaled and noise disturbance. Compared with the classical ICP algorithm, the proposed algorithm improves the registration efficiency of the Bunny point cloud data by 98% and the registration speed of the Dragon point cloud data by at least 20 times. Moreover, in the registration of the large-scale Dragon point cloud data, the registration time of the proposed algorithm is 6210.4 s less than that of the classical ICP algorithm, and the registration accuracy is higher than those of other algorithms. The proposed algorithm does not fall into the local minimum and possesses obvious advantages in terms of fast and accurate registration and stability.
引文
[1] Besl P J, McKay N D. A method for registration of 3-D shapes[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1992, 14(2): 239-256.
[2] He S J, Zhao S T, Bai F, et al. A method for spatial data registration based on PCA-ICP algorithm[J]. Advanced Materials Research, 2013, 718-720: 1033-1036.
[3] Chen C S, Hung Y P, Cheng J B. RANSAC-based DARCES: a new approach to fast automatic registration of partially overlapping range images[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1999, 21(11): 1229-1234.
[4] Sharp G C, Lee S W, Wehe D K. ICP registration using invariant features[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(1): 90-102.
[5] Du S Y, Xu Y T, Wan T, et al. Robust iterative closest point algorithm based on global reference point for rotation invariant registration[J]. PLoS One, 2017, 12(11): e0188039.
[6] Li R Z, Yang M, Tian Y, et al. Point cloud registration algorithm based on the ISS feature points combined with improved ICP algorithm[J]. Laser & Optoelectronics Progress, 2017, 54(11): 111503. 李仁忠, 杨曼, 田瑜, 等. 基于ISS特征点结合改进ICP的点云配准算法[J]. 激光与光电子学进展, 2017, 54(11): 111503.
[7] Yang J L, Li H D, Campbell D, et al. Go-ICP: a globally optimal solution to 3D ICP point-set registration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(11): 2241-2254.
[8] Yang J L, Li H D, Jia Y D. Go-ICP: solving 3D registration efficiently and globally optimally[C]. IEEE International Conference on Computer Vision, 2013: 1457-1464.
[9] Ji S J, Ren Y C, Ji Z, et al. An improved method for registration of point cloud[J]. Optik, 2017, 140: 451-458.
[10] Ying S H, Peng J G, Du S Y, et al. A scale stretch method based on ICP for 3D data registration[J]. IEEE Transactions on Automation Science and Engineering, 2009, 6(3): 559-565.
[11] Li Q S, Xiong R, Vidal-Calleja T. A GMM based uncertainty model for point clouds registration[J]. Robotics and Autonomous Systems, 2017, 91: 349-362.
[12] Jian B, Vemuri B C. Robust point set registration using gaussian mixture models[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(8): 1633-1645.
[13] Zhang Z, Xu H L, Yin H. A fast point cloud registration algorithm based on key point selection[J]. Laser & Optoelectronics Progress, 2017, 54(12): 121001. 张哲, 许宏丽, 尹辉. 一种基于关键点选择的快速点云配准算法[J]. 激光与光电子学进展, 2017, 54(12): 121001.
[14] Prakhya S M, Liu B B, Lin W S, et al. B-SHOT: a binary 3D feature descriptor for fast Keypoint matching on 3D point clouds[J]. Autonomous Robots, 2017, 41(7): 1501-1520.
[15] Ge X M. Automatic markerless registration of point clouds with semantic-keypoint-based 4-points congruent sets[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, 130: 344-357.
[16] Quan S W, Ma J, Hu F Y, et al. Local voxelized structure for 3D binary feature representation and robust registration of point clouds from low-cost sensors[J]. Information Sciences, 2018, 444: 153-171.
[17] Myronenko A, Song X B. Point set registration: coherent point drift[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(12): 2262-2275.
[18] Zhang X, Zhang A W, Wang Z H. Point cloud registration based on improved normal distribution transform algorithm[J]. Laser & Optoelectronics Progress, 2014, 51(4): 041002. 张晓, 张爱武, 王致华. 基于改进正态分布变换算法的点云配准[J]. 激光与光电子学进展, 2014, 51(4): 041002.
[19] Tang Z R, Liu M Z, Wang C, et al. Point cloud registration algorithm based on multi-dimensional mixed Cauchy distribution[J]. Acta Optica Sinica, 2019, 39(1): 0115005. 唐志荣, 刘明哲, 王畅, 等. 基于多维混合柯西分布的点云配准[J]. 光学学报, 2019, 39(1): 0115005.
[2] He S J, Zhao S T, Bai F, et al. A method for spatial data registration based on PCA-ICP algorithm[J]. Advanced Materials Research, 2013, 718-720: 1033-1036.
[3] Chen C S, Hung Y P, Cheng J B. RANSAC-based DARCES: a new approach to fast automatic registration of partially overlapping range images[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1999, 21(11): 1229-1234.
[4] Sharp G C, Lee S W, Wehe D K. ICP registration using invariant features[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(1): 90-102.
[5] Du S Y, Xu Y T, Wan T, et al. Robust iterative closest point algorithm based on global reference point for rotation invariant registration[J]. PLoS One, 2017, 12(11): e0188039.
[6] Li R Z, Yang M, Tian Y, et al. Point cloud registration algorithm based on the ISS feature points combined with improved ICP algorithm[J]. Laser & Optoelectronics Progress, 2017, 54(11): 111503. 李仁忠, 杨曼, 田瑜, 等. 基于ISS特征点结合改进ICP的点云配准算法[J]. 激光与光电子学进展, 2017, 54(11): 111503.
[7] Yang J L, Li H D, Campbell D, et al. Go-ICP: a globally optimal solution to 3D ICP point-set registration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(11): 2241-2254.
[8] Yang J L, Li H D, Jia Y D. Go-ICP: solving 3D registration efficiently and globally optimally[C]. IEEE International Conference on Computer Vision, 2013: 1457-1464.
[9] Ji S J, Ren Y C, Ji Z, et al. An improved method for registration of point cloud[J]. Optik, 2017, 140: 451-458.
[10] Ying S H, Peng J G, Du S Y, et al. A scale stretch method based on ICP for 3D data registration[J]. IEEE Transactions on Automation Science and Engineering, 2009, 6(3): 559-565.
[11] Li Q S, Xiong R, Vidal-Calleja T. A GMM based uncertainty model for point clouds registration[J]. Robotics and Autonomous Systems, 2017, 91: 349-362.
[12] Jian B, Vemuri B C. Robust point set registration using gaussian mixture models[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(8): 1633-1645.
[13] Zhang Z, Xu H L, Yin H. A fast point cloud registration algorithm based on key point selection[J]. Laser & Optoelectronics Progress, 2017, 54(12): 121001. 张哲, 许宏丽, 尹辉. 一种基于关键点选择的快速点云配准算法[J]. 激光与光电子学进展, 2017, 54(12): 121001.
[14] Prakhya S M, Liu B B, Lin W S, et al. B-SHOT: a binary 3D feature descriptor for fast Keypoint matching on 3D point clouds[J]. Autonomous Robots, 2017, 41(7): 1501-1520.
[15] Ge X M. Automatic markerless registration of point clouds with semantic-keypoint-based 4-points congruent sets[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, 130: 344-357.
[16] Quan S W, Ma J, Hu F Y, et al. Local voxelized structure for 3D binary feature representation and robust registration of point clouds from low-cost sensors[J]. Information Sciences, 2018, 444: 153-171.
[17] Myronenko A, Song X B. Point set registration: coherent point drift[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(12): 2262-2275.
[18] Zhang X, Zhang A W, Wang Z H. Point cloud registration based on improved normal distribution transform algorithm[J]. Laser & Optoelectronics Progress, 2014, 51(4): 041002. 张晓, 张爱武, 王致华. 基于改进正态分布变换算法的点云配准[J]. 激光与光电子学进展, 2014, 51(4): 041002.
[19] Tang Z R, Liu M Z, Wang C, et al. Point cloud registration algorithm based on multi-dimensional mixed Cauchy distribution[J]. Acta Optica Sinica, 2019, 39(1): 0115005. 唐志荣, 刘明哲, 王畅, 等. 基于多维混合柯西分布的点云配准[J]. 光学学报, 2019, 39(1): 0115005.