3D Reconstruction Approach for Outdoor Scene Based on Multiple Point Cloud Fusion

被引:10
作者
Chen, Hui [1 ]
Feng, Yan [1 ]
Yang, Jian [1 ]
Cui, Chenggang [1 ]
机构
[1] Shanghai Univ Elect Power, Sch Automat Engn, Shanghai Key Lab Power Stn Automat Technol, Shanghai 200090, Peoples R China
基金
上海市自然科学基金; 中国国家自然科学基金;
关键词
Outdoor scene 3D reconstruction; Point cloud registration; Structure from motion (SfM); Laser scanning; REGISTRATION; 2D;
D O I
10.1007/s12524-019-01029-y
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Multiple point cloud fusion is one of the most widely used methods for outdoor scene 3D reconstruction. However, being based on the traditional registration methods, their performance critically influences the quality of the 3D reconstruction. This paper proposes a 3D reconstruction method that fuses different sensors point cloud, which comes from laser scanning and structure from motion. First, a scale-based principal component analysis-iterative closest point (a scaled PCA-ICP) algorithm is addressed to eliminate different scales of two view points. Further, the feature points are extracted automatically for accurate registration by analyzing the persistence of feature points with discretely sampling on different sphere radii. Finally, the optimization ICP method is used to match multiple point cloud to achieve accurate reconstruction of outdoor scenes robustly. The experimental evaluation demonstrates that the proposed method is able to produce reliable registration results for the outdoor scene.
引用
收藏
页码:1761 / 1772
页数:12
相关论文
共 50 条
[41]   Aligning Algorithm of 3D Point Cloud Model Based on Dimensionality Reduction [J].
He, Lijiang ;
Li, Zhi ;
Chen, Shuqin .
2017 2ND INTERNATIONAL CONFERENCE ON MULTIMEDIA AND IMAGE PROCESSING (ICMIP), 2017, :281-285
[42]   Research on 3D Point Cloud Registration Algorithm Based on FPFH and ColorICP [J].
Pan, Junjie ;
Zhang, Xu .
INTERNATIONAL CONFERENCE ON OPTICAL AND PHOTONIC ENGINEERING, ICOPEN 2024, 2025, 13509
[43]   Automatic dimension inspection of industrial parts based on 3D point cloud [J].
Li, Xudong ;
Yao, Pandeng ;
Jiang, Hongzhi ;
Zhao, Huijie .
SEVENTH INTERNATIONAL CONFERENCE ON OPTICAL AND PHOTONIC ENGINEERING (ICOPEN 2019), 2019, 11205
[44]   FPFH-Based Graph Matching for 3D Point Cloud Registration [J].
Zhao, Jiapeng ;
Li, Chen ;
Tian, Lihua ;
Zhu, Jihua .
TENTH INTERNATIONAL CONFERENCE ON MACHINE VISION (ICMV 2017), 2018, 10696
[45]   Normal direction measurement based on 3D point cloud in robotic drilling [J].
Rao, Gang ;
Yang, Xiangdong ;
Xu, Jing ;
Chen, Ken ;
Zhang, Guanglie .
2017 IEEE 7TH ANNUAL INTERNATIONAL CONFERENCE ON CYBER TECHNOLOGY IN AUTOMATION, CONTROL, AND INTELLIGENT SYSTEMS (CYBER), 2017, :1182-1187
[46]   3D point cloud recognition of substation equipment based on plane detection [J].
Yuan, Qianjin ;
Luo, Yong ;
Wang, HeShan .
RESULTS IN ENGINEERING, 2022, 15
[47]   An Automatic 3D Point Cloud Registration Method Based on Biological Vision [J].
Liu, Jinbo ;
Guo, Pengyu ;
Sun, Xiaoliang .
APPLIED SCIENCES-BASEL, 2021, 11 (10)
[48]   3D POINT CLOUD REGISTRATION WITH SHAPE CONSTRAINT [J].
Agarwal, Swapna ;
Bhowmick, Brojeshwar .
2017 24TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2017, :2199-2203
[49]   3D point cloud registration algorithm with IVCCS [J].
Wang C. ;
Li G. ;
Liu X. ;
Shi C. ;
Qiu W. .
Hongwai yu Jiguang Gongcheng/Infrared and Laser Engineering, 2022, 51 (06)
[50]   An iterative closest point algorithm based on biunique correspondence of point clouds for 3D reconstruction [J].
Wei, Shengbin ;
Wang, Shaoqing ;
Zhou, Changhe ;
Liu, Kun ;
Fan, Xin .
Guangxue Xuebao/Acta Optica Sinica, 2015, 35 (05)