雷射掃描技術已非常普遍，不論利用點雷射來測量距離，或於工業中使用線雷射掃描作為加工成品之精密量測、甚至使用面雷射掃描針對各種物件之逆向工程掃描用以獲得3D空間中的點陣雲檔都已相當成熟。
但若要從掃描之點雲檔進而自動化獲取加工路徑，目前現有的方法都是從3D的點雲資訊，透過演算法，找出特定的特徵點再進行軌跡規劃。而從3D點雲資訊做運算找特徵的缺點是需要花費大量時間進行3D的資訊點運算，且由於物件若是過於複雜將難以找到正確的軌跡。因此本文中將提到如何透過自建多組線雷射掃描系統，並利用新方法拼接多組掃描，且能夠在掃描物件時，從2D影像中，利用影像辨識演算法，找出待加工之特徵點，並轉換成3D的點雲檔。
各組雷射掃描系統由一部相機及一支雷射光源組成，透過事先計算出光平面方程式，再從雷射光投影至物體上所形成的曲線Pixel點(c, r)，將這些Pixel點放進所算出的光平面方程式，算出3D的點雲檔資訊。
拼接多組掃描系統將會運用到「雙面棋盤格」進行相機校正並獲得外部參數，以及「五角金字塔」作為掃描之物件，並透過SVD算出各組掃描系統間的轉換矩陣完成拼接，最後影像演算法之建立運用多項OpenCV功能，作為輔助，可針對不同物件外形之需求，寫出適合的一套2D影像演算法，將特徵軌跡在掃描結束同時即同步完成特徵軌跡生成。

Laser scanning technologies have become very common in the last decade. Modern laser scanning operations which include using point laser scanners to measure distances and using line laser scanners for scanning in the industry as precision measurement of products are quite mature.
To automatically obtain the processing path from the point cloud file, most current existing methods are based on using algorithm to calculate the specific feature points for the trajectory planning based on 3D point cloud information. The corresponding disadvantage of finding features by such operations is the large amount of time required to perform calculations on 3D information points. It will be also relatively difficult to find the correct trajectory if the shape of the object is complex. Therefore, this research will aim to build a multi-set line laser scanning system. While scanning objects, the image recognition algorithms based on 2D images is used to find desired feature points and subsequently convert them into 3D point cloud files of exclusive feature points.
Each set of the laser scanning system is composed of a camera and a laser light source. By calculating the light plane equation in advance, the curve pixel points (c, r) is formed by laser light projected onto the object. These pixel points are put in the calculated light plane equation to calculate the 3D point cloud file information.
This new proposed method for merging multiple point clouds uses the "double-sided chessboard" to calibrate the camera and obtain external parameters, use "pentagonal pyramid" as the scanned object, and then use SVD to calculate the transfer matrix between each point clouds to complete the merging. The establishment of the algorithm will use many OpenCV functions to write required 2D image algorithms for detecting different object shapes, and complete the feature trajectory generation while the object is being scanned.

1. Zhang, H., Ren, Y., Liu, C., & Zhu, J. (2014). Flying spot laser triangulation scanner using lateral synchronization for surface profile precision measurement. Applied optics, 53(20), 4405–4412.
2. https://www.makepartsfast.com/scanner-helps-reverse-engineer-impellor-at-hydroelectric-plant/
3. https://3dwalla.com/3d-scanning.php
4. Ebrahim, M. (2015). Laser Scanners ’ Techniques Overview
5. Tornslev, K. (2005). 3D scanning using multibeam laser.
6. Lee, Ju Hwan & Park, Soon-Yong. (2019). Calibration of Two 3D Sensors with Perpendicular Scanning Directions by using a Piece of Paper. 162-168. 10.5220/0007383301620168.
7. Oomori, S., Nishida, T. & Kurogi, S. Point cloud matching using singular value decomposition. Artif Life Robotics 21, 149–154 (2016)
8. T. Wang, J. Zhang, J. Dong, S. Pan and L. Zheng, "A Method for Generating Spray Trajectory of a Shoe Sole Based on Laser Vision," 2019 IEEE 4th International Conference on Image, Vision and Computing (ICIVC), 2019, pp. 36-39, doi: 10.1109/ICIVC47709.2019.8981040.
9. The MathWorks, I. What Is Camera Calibration? 2018; Available from: http://www.mathworks.com/help/vision/ug/camera-calibration.html?w.mathworks.com.
10. https://www.phyworld.idv.tw/Nature/Jun_2/htm/B3_4-4_POINT.html
11. Duane, C.B., Close-range camera calibration. Photogramm. Eng, 1971. 37(8): p. 855-866.
12. Fryer, J.G. and D.C. Brown, Lens distortion for close-range photogrammetry. Photogrammetric engineering and remote sensing, 1986. 52(1): p. 51-58.
13. Spong, M.W., S. Hutchinson, and M. Vidyasagar, Robot modeling and control. Vol. 3. 2006: Wiley New York.
14. Hartley, R. and A. Zisserman, Multiple view geometry in computer vision. 2003: Cambridge university press.
15. https://docs.opencv.org/3.4/d9/d0c/group__calib3d.html
16. Greco, Marinella & Materazzi, Marzia & Pampaloni, Enrico & Pezzati, Luca & Rocchini, Claudio & Scopigno, Roberto. (2002). Three-dimensional modelling of statues: The Minerva of Arezzo. Journal of Cultural Heritage - J CULT HERIT. 3. 325-331. 10.1016/S1296-2074(02)01242-6.
17. Burger, Wilhelm. (2016). Zhang's Camera Calibration Algorithm: In-Depth Tutorial and Implementation.
18. Le Thai Son , M ,(2016) "Non-contacting 3D Measurement System Integrating Webcam Camera and Laser Projector for Fast Industrial Robot Assembly Implementations"
19. https://docs.opencv.org/3.4/dd/d1a/group__imgproc__feature.html
20. https://docs.opencv.org/3.4/da/d97/tutorial_threshold_inRange.html
21. Zhang, T., & Suen, C. (1984). A fast parallel algorithm for thinning digital patterns. Commun. ACM, 27, 236-239.
22. https://homepages.inf.ed.ac.uk/rbf/HIPR2/thin.htm
23. https://homepages.inf.ed.ac.uk/rbf/HIPR2/gsmooth.htm
24. https://docs.opencv.org/4.5.2/d4/d13/tutorial_py_filtering.html
25. Li, Y., Zhou, J., Huang, F., & Liu, L. (2017). Sub-Pixel Extraction of Laser Stripe Center Using an Improved Gray-Gravity Method †. Sensors, 17(4), 814. MDPI AG. Retrieved from http://dx.doi.org/10.3390/s17040814
26. Usamentiaga, R., & García, D. (2019). Multi-camera calibration for accurate geometric measurements in industrial environments. Measurement, 134, 345-358.
27. Wu, X., Li, Z., & Wen, P. (2017). An automatic shoe-groove feature extraction method based on robot and structural laser scanning. International Journal of Advanced Robotic Systems. https://doi.org/10.1177/1729881416678135