Acta Optica Sinica, Volume. 38, Issue 12, 1215005(2018)
3D Point Cloud Registration Algorithm Based on Feature Matching
Figures & Tables(14)
Fig. 2. Characteristic diagram. (a) Empty circle characteristics; (b) maximized minimum angle characteristics
Fig. 3. Generation process diagram. (a) Insert a new node P; (b) empty circumscribed circle detection; (c) delete the edge AB; (d) form triangles
Fig. 5. Registration charts of a seat. (a) Original point cloud; (b) point cloud of seat; (c) traditional algorithm registration; (d) proposed algorithm registration
Fig. 6. Contrast diagrams of registration time. (a) Total registration time; (b) initial registration time
Fig. 7. Registration charts of car model. (a) Point cloud data; (b) 3D reconstruction diagram; (c) side initial registration chart; (d) side accurate registration chart; (e) top initial registration chart; (f) top accurate registration chart
Fig. 8. Threshold influence curves. (a) Relation curves between average error distance and threshold; (b) relation curves between accurate registration time and threshold
Table 1. Average registration times
View tableTable 1. Average registration times
Algorithm Average total registration time Average initial registration time Average accurate registration time Traditional algorithm 27.055 26.938 0.117 Proposed algorithm 25.511 25.416 0.095
Table 2. Comparison of average error distance and total registration time of multiple registration experiments
View tableTable 2. Comparison of average error distance and total registration time of multiple registration experiments
Group Average registration error of traditional algorithm /cm Average registration error of proposed algorithm /cm Total registration time of traditional algorithm /s Total registration time of proposed algorithm /s 1 0.337 0.301 27.055 25.511 2 0.339 0.301 27.051 25.505 3 0.339 0.303 27.049 25.501 4 0.343 0.311 27.058 25.511 5 0.348 0.316 27.056 25.509 6 0.338 0.307 27.052 25.507 7 0.335 0.301 27.063 25.524 8 0.340 0.306 27.046 25.497
Table 3. Experimental parameters
View tableTable 3. Experimental parameters
Number of point cloud ICP accurate registration parameter Threshold Maximum number of iterations Transform matrix difference Mean square error Source point cloud 25717 0.01 500 1×10-10 0.1 Target point cloud 50887
Table 4. Point cloud conversion results of different algorithms
View tableTable 4. Point cloud conversion results of different algorithms
Algorithm Rotational translation matrix of initial registration Rotational translation matrix of accurate registration Traditional algorithm Proposed algorithm
Table 5. [in Chinese]
View tableTable 5. [in Chinese]
Algorithm Average total registration time Average initial registration time Average accurate registration time Traditional algorithm 46.358 46.210 0.148 Proposed algorithm 45.033 44.892 0.141
Table 6. Experimental results of different algorithms
View tableTable 6. Experimental results of different algorithms
Algorithm X -direction rotationangle /rad Y -direction rotationangle /rad Z -direction rotationangle /rad Average error distance /cm Traditional algorithm 0.038 0.564 0.291 0.564 Proposed algorithm 0.023 0.534 0.277 0.533
Tools
Get Citation
Copy Citation Text
Jian Liu, Di Bai. 3D Point Cloud Registration Algorithm Based on Feature Matching[J]. Acta Optica Sinica, 2018, 38(12): 1215005
Paper Information
Category: Machine Vision
Received: Jun. 4, 2018
Accepted: Jul. 26, 2018
Published Online: May. 10, 2019
The Author Email:
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20