Point Cloud Registration Method Based on 2D Image Feature

doi:10.13474/j.cnki.11-2246.2017.0313

Abstract

Abstract: A point cloud registration method is proposed in the paper in order to improve the registration accuracy and convergence rate of low overlapping point clouds.Firstly,a point cloud registration algorithm based on region level is used to complete coarse registration;Secondly,3D point cloud is converted into 2D image,SURF(speeded up robust features)algorithm is used to extract 2D image features,and the matching pixel pairs are solved;Finally,3D corresponding points are gotten according to the 2D matching pixels,and the rigid transformation is solved,then the fast and accurate registration of point cloud are achieved.The experimental results showed that the registration accuracy and convergence rate of proposed point cloud registration method improved about 20% and 60% respectively compared with iterative closest point(ICP)algorithm,it is a high accurate and fast point cloud registration method.

Key words: point cloud registration, image feature, region level, rotation matrix, iterative closest point

CLC Number:

ZHAO Fuqun, ZHOU Mingquan. Point Cloud Registration Method Based on 2D Image Feature[J]. 测绘通报, 2017, 0(10): 39-42,73.

References

[1] LUO Y,GUAN T,WEI B,et al.Fast Terrain Mapping from Low Altitude Digital Imagery[J].Neurocomputing,2015(156):105-116.
[2] 闫阳阳,李永强,王英杰,等.三维激光点云联合无人机影像的三维场景重建研究[J].测绘通报,2016(1):84-87.
[3] 郑敏辉,臧玉府,梁福逊,等.不同场景的地面激光点云配准方法研究[J].测绘通报,2015(8):30-34.
[4] GAO Y,WANG M,TAO D,et al. 3-D Object Retrieval and Recognition with Hypergraph Analysis[J].IEEE Transactions on Image Processing,2012,21(9):4290-4303.
[5] GAO Y,WANG M,JI R,et al. 3-D Object Retrieval with Hausdorff Distance Learning[J].IEEE Transactions on Industrial Electronics,2014,61(4):2088-2098.
[6] 税午阳,周明全,武仲科,等.数据配准的颅骨面貌复原方法[J].计算机辅助设计与图形学学报,2011,23(4):607-614.
[7] BESL P J,MCKAY N D.A Method for Registration of 3-D Shapes[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2002,14(2):239-256.
[8] ZHU J,DU S,YUAN Z,et al.Robust Affine Iterative Closest Point Algorithm with Bidirectional Distance[J].Iet Computer Vision,2012,6(3):252-261.
[9] HAN J,YIN P,HE Y,et al.Enhanced ICP for the Registration of Large-scale 3D Environment Models:An Experimental Study[J].Sensors,2016,16(2):1-15.
[10] DU S,LIU J,ZHANG C.Probability Iterative Closest Point Algorithm for m-D Point Set Registration with Noise[J].Neurocomputing,2015(157):187-198.
[11] DU S,ZHENG N,XIONG L.Scaling Iterative Closest Point Algorithm for Registration of m-D Point Sets[J].Journal of Visual Communication and Image Representation,2010,21(5-6):442-452.
[12] ZHAO L,SHEN X,LONG X.Robust Wrinkle-aware Non-rigid Registration for Triangle Meshes of Hand with Rich and Dynamic Details[J].Computers & Graphics,2012,36(5):577-583.
[13] 戴玉成,张爱武.三维激光扫描数据快速配准算法研究[J].测绘通报,2010(6):8-11.
[14] 韩宝昌,曹俊杰,苏志勋.一种区域层次上的自动点云配准算法[J].计算机辅助设计与图形学学报,2015,27(2):313-319.
[15] HORN B K P.Closed-form Solution of Absolute Orientation Using Unit Quarternions[J].Journal of the Optical Society of American, 1987,4(4):629-642.
[16] SCARAMUZZA D,HARATI A,SIEGWART R.Extrinsic Self Calibration of a Camera and a 3D Laser Range Finder from Natural Scenes[C]//Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems.San Diego:IEEE,2007:4164-4169.
[17] BAY H,ESS A,TUYTELAARS T,et al.Speeded-up Robust Features(SURF).Computer Vision and Image Understanding,2008,110(3),346-359.
[18] Bay H,Tuytelaars T,VAN GOOL L.SURF:Speeded Up Robust Features[C]//Computer Vision-ECCV 2006.Berlin:Springer,2006:404-417.
[19] Stanford Computer Graphics Laboratory.The Stanford 3D Scanning Repository[EB/OL].(1996-9-10)[2016-06-15].http://graphics.stanford.edu/data/3Dscanrep.