• Laser & Optoelectronics Progress
  • Vol. 60, Issue 14, 1428008 (2023)
Yi Chen, Yong Wang*, Jinlong Li, Dengzhi Liu, Xiaorong Gao, and Yu Zhang
Author Affiliations
  • School of Physical Science and Technology, Southwest Jiaotong University, Chengdu 610031, Sichuan, China
  • show less
    DOI: 10.3788/LOP222075 Cite this Article Set citation alerts
    Yi Chen, Yong Wang, Jinlong Li, Dengzhi Liu, Xiaorong Gao, Yu Zhang. An Efficient Point Cloud Registration Algorithm Based on Principal Component Analysis[J]. Laser & Optoelectronics Progress, 2023, 60(14): 1428008 Copy Citation Text show less
    Stanford dataset. (a) bunny model; (b) armadillo model; (c) bunny registration initial pose; (d) armadillo registration initial pose
    Fig. 1. Stanford dataset. (a) bunny model; (b) armadillo model; (c) bunny registration initial pose; (d) armadillo registration initial pose
    Registration of bunny. (a) SAC-IA algorithm; (b) 4PCS algorithm; (c) PCA algorithm; (d) improved PCA algorithm
    Fig. 2. Registration of bunny. (a) SAC-IA algorithm; (b) 4PCS algorithm; (c) PCA algorithm; (d) improved PCA algorithm
    Registration of armadillo. (a) SAC-IA algorithm; (b) 4PCS algorithm; (c) PCA algorithm; (d) improved PCA algorithm
    Fig. 3. Registration of armadillo. (a) SAC-IA algorithm; (b) 4PCS algorithm; (c) PCA algorithm; (d) improved PCA algorithm
    Train components. (a) Standard rail model; (b) bolt point cloud after denoising; (c) brake cylinder; (d) gearbox
    Fig. 4. Train components. (a) Standard rail model; (b) bolt point cloud after denoising; (c) brake cylinder; (d) gearbox
    Registration results of proposed algorithm for train components. (a)-(d) Initial positions of point clouds of each component; (e)-(h) coarse registration results by improved PCA algorithm
    Fig. 5. Registration results of proposed algorithm for train components. (a)-(d) Initial positions of point clouds of each component; (e)-(h) coarse registration results by improved PCA algorithm
    Registration of rail model by 4PCS algorithm. (a) 4PCS coarse registration results; (b) ICP fine registration results
    Fig. 6. Registration of rail model by 4PCS algorithm. (a) 4PCS coarse registration results; (b) ICP fine registration results
    Point cloud data of sand-spreading pipe. (a) Point cloud data of sand pipe after discrete points removed; (b) initial pose
    Fig. 7. Point cloud data of sand-spreading pipe. (a) Point cloud data of sand pipe after discrete points removed; (b) initial pose
    Point cloud registration effect of sand-spreading pipe. (a)-(d) Coarse registration results of SAC-IA, 4PCS, PCA, and improved PCA respectively; (e)-(h) results of Fig.8 (a)-(d) by ICP fine registration
    Fig. 8. Point cloud registration effect of sand-spreading pipe. (a)-(d) Coarse registration results of SAC-IA, 4PCS, PCA, and improved PCA respectively; (e)-(h) results of Fig.8 (a)-(d) by ICP fine registration
    Point cloud data of brake cylinder. (a) Brake cylinder 1; (b) two point clouds collected from perspective 1; (c) brake cylinder 2; (d) two point clouds collected from perspective 2
    Fig. 9. Point cloud data of brake cylinder. (a) Brake cylinder 1; (b) two point clouds collected from perspective 1; (c) brake cylinder 2; (d) two point clouds collected from perspective 2
    Registration of two point clouds from perspective of brake cylinder 1. (a)-(d) coarse registration results of SAC-IA, 4PCS, PCA, and improved PCA respectively
    Fig. 10. Registration of two point clouds from perspective of brake cylinder 1. (a)-(d) coarse registration results of SAC-IA, 4PCS, PCA, and improved PCA respectively
    Registration of two point clouds from perspective of brake cylinder 2. (a)-(d) coarse registration results of SAC-IA, 4PCS, PCA, and improved PCA respectively
    Fig. 11. Registration of two point clouds from perspective of brake cylinder 2. (a)-(d) coarse registration results of SAC-IA, 4PCS, PCA, and improved PCA respectively
    bunnyΔRΔTRRMSEt /s
    SAC-IA16.57390.05306080.002932811.29
    4PCS15.02420.05482990.001111641.659
    PCA0.0001263587.57978×10-75.27204×10-723.873
    Improved PCA0.0001263587.57978×10-75.27204×10-72.571
    Table 1. Experimental indicators of bunny registration
    ArmadilloΔRΔTRRMSEt /s
    SAC-IA4.714424.040721.509612.131
    4PCS1.090639.869021.959824.9
    PCA0.0001229430.000361710.00041345627.007
    Improved PCA0.0001229430.000361710.0004134562.751
    Table 2. Experimental indicators of armadillo registration
    Train componentsΔRΔTRRMSEt /s
    Rail0.0002732080.008022710.006983713.511
    Bolt0.0001058680.001936860.00161995.329
    Brake cylinder8.19623e-50.002184110.0009876442.03
    Gearbox0.0001229430.000361710.0004134562.751
    Table 3. Experimental indicators of train components registration
    t /sSAC-IA4PCSPCAImproved PCA
    Rail98.8371.79111.0663.511
    Bolt6.2097.14823.9765.329
    Brake cylinder41.1564.0923.3252.03
    Gearbox22.0191.43522.9482.751
    Table 4. Comparison of registration time of train components
    AlgorithmΔRΔTRRMSEt /s
    SAC-IA3.9515533.04661.639445.37
    4PCS1.2009216.98891.288895.656
    PCA0.2669652.007670.52207823.934
    Improved PCA0.0001092830.0004488350.0006059294.427
    SAC-IA+ICP0.1947833.175720.4099671.103
    4PCS+ICP0.64949616.10110.6539281.376
    PCA+ICP0.004644530.0260820.307260.682
    Imp-PCA+ICP0.0003415090.002011960.0009956850.464
    Table 5. Experimental indicators of sand-spreading pipe registration
    t /sSAC-IA4PCSPCAImproved PCA
    Brake cylinder 127.4341.72223.3934.939
    Brake cylinder 226.91148.52596.7481.34
    Table 6. Coarse registration time of brake cylinder
    Yi Chen, Yong Wang, Jinlong Li, Dengzhi Liu, Xiaorong Gao, Yu Zhang. An Efficient Point Cloud Registration Algorithm Based on Principal Component Analysis[J]. Laser & Optoelectronics Progress, 2023, 60(14): 1428008
    Download Citation