透過您的圖書館登入
IP:18.117.183.150
  • 學位論文

使用Kinects掃描人體實現三維重建與尺寸量測之系統

System for Implementing Three-dimensional reconstruction and Size Measurement of Human Body by using Kinects

指導教授 : 黃榮堂 蔡定江
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


以往大部分的人體三維掃描與重建系統,多數是以雷射掃描儀進行人體掃描重建。雖然雷射掃描儀有高精準度但是價格昂貴及使用上不方便,必須有專業的周邊設備,而且必須經過一定的訓練才可以操作。 本研究主旨是希望設計一種低成本簡易操作,以及精準度近似雷射掃描儀的三維重建與尺寸量測系統。故本研究使用低成本的掃描裝置,例如KINECT,進行人體掃描。其中因KINECT掃描視角有限,無法在有限的空間內掃描全身和擷取到足夠多的雲點。KINECT為固定方位掃描,不能全方位的掃描全身,若想達成全方位的掃描就需要多方位架設KINECT向內進行掃描,但這樣的設計會大幅增加成本。綜合以上所述,本研究提出一種系統,利用上下兩台KINECT和旋轉平台連結電腦做掃描與控制。而上下兩台KINECT可在有限的空間內掃描全身以及擷取到足夠的雲點,並且結合旋轉平台使人體自轉一圈以達到全方位的掃描。KINECT掃描多張不同旋轉角度下的身體雲點,接合成完整身體三維尺寸座標,此三維尺寸座標可以提供身體的各部分的尺寸量測。

並列摘要


In the past, most of the three-dimensional body scanning and reconstruction systems are using Laser scanner to do Body scanning. The laser scanner is highly precise, but expensive and not convenient. In addition, it must have professional peripherals, and is required by an operator with training. The purpose of this study is to design one kind of three-dimensional reconstruction and measurement system which is low-cost, simple and approximate accuracy to the laser scanner. One of possible choice is RGB-depth camera, such as Kinect. However, one single Kinect is unable to scan whole body and capture enough of the cloud point due to its limitation of scanning perspective. It is necessary to set up multi-faceted KINECT to reach a full range of scanning, but this way will increase in costs significantly. To sum up, this study proposes a system by using two Kinects, one upper and one lower, and a rotating platform linking the computer to do the scanning and control. By this way, it can scan whole body and capture enough of the cloud point in the limited room. Moreover the combination of rotating platform can achieve a full range of scanning by scanning from different angle. In the last, combination of all the pictures cans reconstruct the three-dimensional body, which then can provide size information for each parts of the body.

參考文獻


[6] 林詩婷,使用單一Kinect相機及時建構三維紋理模型,碩士論文,國立中正大學資訊工程研究所,民國101年2月。
[8] Sean Kean, Jonathan Hall, Phoenix Perry, “Meet the Kinect An Introduction to Programming Natural User Interfaces, ” Apress , 2012.
[13] Nicolas Burrus, Kinect Calibration
[14] Rafael C. Gonzalez, Richard E. Woods,Digital Image Processing (3rd Edition),Prentice Hall,2007-08-21
[16] Yoav Freund and Robert E. Schapire. A decision-theoretic generalization of on-line learning and an application to boosting. Journal of Computer and System Sciences, 55(1):119–139,August 1997.

延伸閱讀