多維度結構光線圖案應用在三維介面掃描

Multi-Dimension Structural Light Patterns for 3D Surface Scanning

施賢孝(SHIH-HSIEN HSIAO)

指導教授 ： 鄭文凱

國立暨南國際大學/科技學院/通訊工程研究所/碩士(2008年)

https://doi.org/10.6837/NCNU.2008.00030

摘要

三維掃描系統在很多工業方面都有被運用的身影， Gray code是比較常被運用在拍攝上的編碼。本論文針對現有Gray code編碼的缺點進行研究與改良。
Gray code有bit變換頻率集中在同一列的問題，容易產生投影的影像強度互相干擾，造成解碼時量化錯誤。而Gray code的相鄰codewords的Euclidean distance太小，解碼時量化錯誤一定會產生誤判到別的codewords。
本論文透過Error Correcting Code (ECC)方式解決相鄰codewords的Euclidean distance太小問題，也透過設定bit重複長度編出bit變換頻率平均分散各列的編碼，從實驗數據證明針對Gray code缺點改良的編碼方式都可以提升編碼效能。
我們在研究過程發現投射高影像強度光源容易造成拍攝環境變亮，我們稱為亮暗互擾，影響攝影機接收到的影像強度值。我們透過提高量化門檻值或是投射低影像強度值的方式避免亮暗互擾的影響，套用到Gray code上，證實有明顯的效能提升。

關鍵字

三維掃描系統 ； 相鄰codewords ； 亮暗互擾 ； 量化門檻值

並列摘要

Recently the 3D scanner technology has become a significant issue in computer vision fields and has been widely applied. In the technology, Gray code is the code usually used. The Euclidean distance of the adjacent codewords in Grey code is, however, too small, and the change of bit values is concentrated on the same row, and hence decoding errors are easily to happen. In this thesis, a new approach based on error correcting code (ECC), whose distance property is better than that of Gray code, has been proposed. Also, the change of bit values in the proposed codes is found equally distributed on all different rows.
Moreover, a phenomenon, defined as "bright-dark interference", is examined in the study. This problem is caused by the interference between the different intensities of the light patterns projected on the scanned project. The solution of the problem is to increase the threshold of the quantification on decoding, or to decrease the intensity of the entire light patterns. The experiment results show that the accuracy of the 3D scanning has been improved in the proposed strategies, compared to the conventional design on Grey code.

並列關鍵字

3D scanner ； adjacent codewords ； bright-dark interference ； Euclidean distance

參考文獻

[1] G. Sansoni, S. Corini, S. Lazzari, R. Rosedda, F. Docchio, “3D Imaging based on Gray code light projection : Characterization of the measuring algorithm and development of a measuring system for industrial application,” Applied Optics, Vol. 36, pp.4463-4472, 1997.

[2] M. Rioux, “Laser range finder based on synchronized scanners,” Applied Optics, Vol. 23, pp.3837-3843,1984.

[3] T. G.. Stahs, F. M. Wahl, “Fast and robust range data acquisition in a low-cast environment,” Proceeding of the ISPRS-Conference: Close-Range Photogrammetry meets Machine Vision, 1395, SPIE, pp.496-503, 1990.

[4] R.G. Chen, “Complete Object Surface Scanning with Camera Pose Estimation,” Master’s thesis, NCNU, July 2003.

[5] J. Weng, P. Cohen, M Herniou, “Camera Calibration with distortion models and accuracy evaluation,” IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol. 14, No. 10, pp.965-980, 1992.