透過您的圖書館登入
IP:13.59.61.119
  • 學位論文

具大量測範圍彩色深度影像視覺系統之研發

Development of RGB-D imaging system with a high range of measuring depth

指導教授 : 陳亮嘉
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


本論文旨主要是開發新的一套機器視覺量測系統,目前設定在室內環境下進行大範圍且遠距離三維視覺量測,利用微型投影機將所設計之週期性條紋結合特徵點結構光投射至量測範圍空間中,並可安裝於移動型機器人上擷取到全域式的即時影像,未來可整合有效的物件分離與判斷技術,提高視覺系統可靠度與自我決策之能力,這將會對目前使用主動式量測視覺技術有相當大的提升和創新。 本研究亦利用影像之二維頻譜分 佈推導傅立葉演算法之量測法則,可將投影之單張影像建立待測物三維影像重建之結果,並結合投影特徵點結合對映函數法則來判斷待測物有效三維重建影像之絕對距離,將可解決傳統參考平面之相對座標之量測限制,可改善使用傅立葉週期性條紋量測之無參考背景情況進行物體形貌判斷,其投影至空間中多重特徵點進行多重深度物體之絕對深度判斷基準,且量測範圍可達六公尺遠,並將深度影像結合彩色攝影機所擷取之彩色影像資訊,達到大範圍彩色深度影像自動化物體形貌量測之效果。

並列摘要


FTP is one of the leading techniques which can provide accurate 3-D profile of an object surface; however, the weakness of FTP method is the dependence on reference plane to retrieve the phase difference and depth value. The depth obtained from triangulation principle is actually the relative depth between the points on the object surface and the defined reference plane. Therefore, the position of reference plane is very important to define and locate the 3-D object into the world coordinate. This prevents the FTP method from being integrated within any hand-hold device for acquiring 3-D images. This research overcomes this weakness of FTP method and develop a compact 3-D image acquisition system which can be freely moved during operation and maintain high accuracy of 3-D depth measurement. An set of extra reference points is added as the absolute reference points for detecting the absolute distance. The image acquisition unit is an near-infrared CCD camera being connected with a portable notebook computer. Before doing any measurement, the system is calibrated through a system calibration process to obtain the mapping functions for the projecting reference points. These calibrated functions can be employed during further image acquisition process for calculating the absolute distance between these reference points and the detected 3-D scanner.

參考文獻


[37 ]何宣緯,具高精度與段高量測範圍傅立葉轉換三維形貌量測系統之研發,碩士論文,國立臺北科技大學自動化科技研究所,臺北,2008。
[1 ]R. Lange and P. Seitz, "Solid-State Time-of-Flight Range Camera", IEEE Journal of quantum electronics, Vol. 37, 2001.
[2 ]L. Guan, J. S. Franco, M. Pollefeys, "3D Object reconstruction with heterogeneous sensor data" Proceedings of 3DPVT'08 - the Fourth International Symposium on 3D Data Processing, Visualization and Transmission, Georgia Institute of Technology, Atlanta, GA, USA, 2008 .
[4 ]H. Fan, G. Li , L. Dong, "The Calibration Algorithm Between 2D Laser Range Finder and Platform", IEEE computer society, 2009.
[7 ] M. Quigley, S. Batra, S. Gould, and E. Klingbeil."High-Accuracy 3D Sensing for Mobile Manipulation Improving Object Detection and Door Opening"IEEE International Conference on Robotics and Automation, 2010.

延伸閱讀