透過您的圖書館登入
IP:3.144.161.116
  • 學位論文

多型樣式數位相移條紋投影術應用於三維多樣表面輪廓量測

Multi-pattern digital phase-shifting fringe projection for three-dimensional various surface profilometry

指導教授 : 劉承揚

摘要


本論文使用三種波形(餘弦波、方波、三角波)的數位條紋結構光,再加上七步相位移法、相位展開技術、參考平面扣除法來量測不同的三維表面輪廓,量測試件包含半圓形、內凹半圓、方形、三角形等形狀。量測系統是將數位結構光法投影至待測物上,然後以數位攝影機擷取條紋影像,最後使用多步相位移技術計算出物體表面的三維輪廓。實驗結果顯示,不同波形的結構光在量測不同樣式的待測物時會達到不同的效果。餘弦波對於半圓形試件有較小的量測誤差(1.27%),方波則對於方形試件有較小的量測誤差(15.66%),而三角波則對於三角形試件有較小的量測誤差(10.18%)。未來應用於產品的即時檢測時,可針對輪廓形狀來選擇較適合的條紋結構光進行量測,以獲得最小誤差的三維表面輪廓。

並列摘要


In this thesis, different three-dimensional surfaces were measured by three kinds of waveforms (cosine wave, square wave, triangular wave) of digital structured light projection, seven-step phase shifting, phase unwrapping, reference plane subtraction method. The measured objects include semi-circular, concave semi-circular, square and triangle. In this system, digital structured light is projected onto the objects, and then the fringe pattern images are captured with CCD. Finally, the three-dimensional surfaces of the objects are calculated by multi-step phase-shifting. As the results, different objects achieved different effect by using different kinds of waveforms of digital structured light. The semi-circular objects measured by cosine wave have smaller measurement error (1.27%). The square objects measured by square wave have smaller measurement error (15.66%). The triangular objects measured by triangular wave have smaller measurement error (10.18%). In order to get the minimum error of the three-dimensional surfaces profile, it is selected more appropriate fringe pattern structured light to measure according to objects profile, which is used for product immediate detection.

參考文獻


[41] 顏子評,“多步相位移干涉術應用於三維複雜曲面量測”, 淡江大學機械與機電工程學系碩士班學位論文(pp. 1-121), 2015
[2]B. Bidanda, S. Motavalli,and K. Harding, “Reverse engineering: an evaluation ofprospective non-contact technologies and application in manufacturing system,”International Journal of Computer Integrated Manufacturing, Vol.4, No. 3,pp.145-156, 1991.
[3] P. J. Besl, “Active optical range imaging sensors,” Machine Vision and Applications, Vol. 1, No. 2, pp.127-152, 1988.
[4]Z. Zhang, S. Huang, Y. Xu, C Chen, Y. Zhao, N. Gao, andY. Xiao, “3D palmprint and hand imaging system based on full-field composite color sinusoidal fringe projection technique,” Applied Optics, Vol. 52, No. 25, pp. 6138-6145, 2013.
[5] S. Huang, Z. Zhang, Y. Zhao, J. Dai, C. Chen, Y. Xu, E. Zhang, and L. Xie, “3Dfingerprint imaging system based on full-field fringe projection profilometry,” Optics and Lasers in Engineering, Vol. 52, pp. 123-130, 2014.

延伸閱讀