簡易檢索 / 詳目顯示

回結果列表
研究生： 張芷瑋
Chang, Chih-wei
論文名稱： 產生有效立體影像之未匹配影像對校正
Rectification of Unmatched Image Pairs for Successful Stereo Image Generation
指導教授： 楊家輝
Yang, Jar-Ferr
學位類別： 碩士
Master
系所名稱： 電機資訊學院 - 電腦與通信工程研究所
Institute of Computer & Communication Engineering
論文出版年： 2007
畢業學年度： 95
語文別： 中文
論文頁數： 85
中文關鍵詞： 立體影像產生 、立體影像校正 、垂直視差 、水平視差
外文關鍵詞： stereo image generation, stereo calibration, vertical parallax, horizontal parallax
相關次數： 點閱：66  下載：8 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 隨著立體液晶螢幕的技術逐漸進步，人們將不再需要配戴立體眼鏡即可透過螢幕欣賞立體影像。然而，對一般使用者而言，立體影像拍攝技術之複雜度較為困難，造成立體影像取得來源並不像一般二維影像普遍。因此，在本論文中，我們提出了一套全自動的立體影像校正方法，以直接將單台相機所拍攝的兩張不同位置角度之影像轉換成為一組成功的立體影像對。
    提出的校正流程中包含：特徵點擷取、雙向對應點配對、相對距離之對應點除錯、影像轉換、裁切與修補影像空洞之步驟。我們將透過這些方法調整未匹配影像對的水平與垂直視差關係，以符合立體視覺。根據實驗結果證明，本論文所提出之校正方法確實可以有效的調整水平與垂直視差，以產生一組適合於立體液晶螢幕觀賞之圖片。

    After availability of 3D LCD display systems, people can perceive stereo images without wearing any special 3D glasses. However, for general family, stereo images are more complicated in photographing, so the sources of stereo images are not as general as 2D images. Therefore, in this thesis, we propose an automatic stereo image calibrating method to rectify two images, which are taken at different view angles, to become a successful pair of stereo image.
    The proposed calibration procedure includes feature point extraction, bidirectional feature point matching, relative distance checking, image transformation, cropping and hole-filling. We will utilize the proposed procedure to adjust horizontal and vertical parallax of general unmatched image pairs to show proper stereo vision.
    Experimental results show that the proposed generation system can effectively adjust vertical and horizontal parallax such that the calibrated stereo image pairs can properly exhibit stereo scenes in stereoscopic LCD display systems.

    摘要(中文) ...................................i Abstract (English) .......................... ii 誌 謝.......................................iii 目 錄.......................................iv 表目錄......................................vii 圖目錄......................................vii 第1章 簡介...................................1 1.1 研究背景.............................................................................1 1.2 立體影像之簡介.................................................................2 1.3 研究目的與動機.................................................................4 1.4 論文大綱.............................................................................6 第2章 立體影像原理及相關校正方法.............7 2.1 立體影像拍攝原理.............................................................7 2.1.1 標準立體影像拍攝方式.................................................................7 2.1.2 立體雙眼相機構造.........................................................................9 2.2 立體影像顯示技術...........................................................11 2.2.1 立體眼鏡.......................................................................................12 2.2.2 裸眼立體螢幕.........................................................................13 2.3 常見之立體影像校正方法...............................................14 2.3.1 二維齊次座標介紹.................................................................14 2.3.2 透過基礎矩陣校正立體影像.................................................17 第3章 立體影像校正原理與演算法設計..........20 3.1 影像校正想法說明...........................................................21 3.2 立體校正演算法之流程設計...........................................24 3.3 特徵點擷取.......................................................................27 3.4 雙向對應點配對...............................................................29 3.5 影像校正轉換...................................................................32 3.5.1 三維齊次座標表示.................................................................33 3.5.2 標準立體影像之深入探討分析.............................................36 3.5.3 轉換矩陣參數設計.................................................................37 3.5.4 轉換參數估算與影像校正轉換.............................................44 3.6 影像裁切與填補...............................................................46 3.6.1 影像邊緣裁切.........................................................................47 3.6.2 影像空洞填補.........................................................................48 第4章 立體影像校正之優化設計................51 4.1 特徵點擷取之前後處理...................................................52 4.1.1 特徵點擷取之前處理—物體輪廓偵測.................................53 4.1.2 特徵點擷取之前處理—主要線段擷取.................................54 4.1.3 特徵點擷取之後處理—特徵點均勻分佈限制.....................56 4.2 對應點配對之後處理─相對距離之對應點除錯...........59 4.3 影像轉換校正—影像旋轉中心設定...............................62 4.4 視覺舒適度—水平視差距離調整...................................65 4.5 演算法流程整理...............................................................67 第5章 實驗結果之分析比較....................71 5.1 實驗步驟結果呈現...........................................................71 5.2 對應點正確率比較...........................................................77 5.3 垂直視差之方均根值比較...............................................78 第6章 系統未來展望與結論....................81 6.1 結論...................................................................................81 6.2 系統未來展望...................................................................81 參考文獻....................................83 表目錄 表（5.1）正確對應點比例....................................................................77 表（5.2）垂直視差之方均根值比較....................................................79 圖目錄 圖（1.1）立體影像成像關係圖…...…………..................…………….3 圖（1.2）立體影像成像位置圖…...…………..................…………….3 圖（2.1）眼球視差圖............................................................................. 8 圖（2.2）立體影像拍攝之平面示意圖..................................................8 圖（2.3）平行立體雙眼相機拍攝示意圖............................................10 圖（2.4）二維平面平移圖....................................................................15 圖（2.5）二維平面旋轉圖....................................................................16 圖（2.6）二維平面縮放圖....................................................................16 圖（2.7）極線幾何關係........................................................................17 圖（3.1）標準立體成像關係圖............................................................21 圖（3.2）標準立體影像........................................................................22 圖（3.3）未校正之立體成像關係圖....................................................23 圖（3.4）立體影像校正流程圖............................................................25 圖（3.5）Sobel 水平垂直運算子.........................................................28 圖（3.6）利用Sobel 運算子求梯度值................................................28 圖（3.7）邊緣鈍化圖............................................................................30 圖（3.8）三維平移轉換圖....................................................................34 圖（3.9）三維X軸旋轉圖...................................................................34 圖（3.10）三維Y軸旋轉圖.................................................................34 圖（3.11）三維Z軸旋轉圖..................................................................35 圖（3.12）三維縮放圖..........................................................................35 圖（3.13）平移旋轉估算示意圖..........................................................40 圖（3.14）以縮放轉換調整深度示意圖..............................................43 圖（3.15）影像邊緣資訊內容遺失圖..................................................47 圖（3.16）影像內部資訊內容遺失圖..................................................49 圖（4.1）影像區域分割圖....................................................................57 圖（4.2）對應點之間的相對關係........................................................60 圖（4.3）垂直高度不同之立體圖旋轉................................................63 圖（4.4）影像旋轉中心平移設定........................................................64 圖（4.5）優化之立體影像校正流程圖................................................68 圖（5.1）四組未校正之影像對............................................................72 圖（5.2）立體影像校正流程步驟結果................................................73 圖（5.3）實驗圖片＃1之校正前後立體影像合成圖比較.................75 圖（5.4）實驗圖片＃2~#4之校正前後立體影像合成圖比較...........76

    參考文獻

    [1] Charles Loop, Zhengyou Zhang, “Computing rectifying homographies for stereo vision”. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Vol.1, pp. 125–131, April 1999

    [2] H.-H.P. Wu, Y.-H. Yu and W.-C. Chen, “Projective rectification based on relative modification and size extension for stereo image pairs,” IEE Proc.-Vis. Image Signal Process., Vol. 152, No. 5, October 2005

    [3] J. Weng, N. Ahuja, T.S. Huang “Matching Two Perspective Views,” IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol. 14, No.8, August 1992.

    [4] William K. Pratt. “Digital Image Processing,” Second Edition, John Wiley and Sons, p.514, 1991.

    [5] C. Harris and M. Stephens, “A combined corner and edge detector,” 4th ALVEY vision conference, pp. 147-151, 1988.

    [6] Levenberg, K. "A Method for the Solution of Certain Problems in Least Squares," Quart. Appl. Math. 2, pp.164-168, 1944.

    [7] Marquardt, D., "An Algorithm for Least-Squares Estimation of Nonlinear Parameters," SIAM J. Appl. Math. 11, pp.431-441, 1963.

    [8] http://www.uart.org.tw/pscl/3d-stereo/3D-Doc-3.htm

    [9] Liang Zhang and Wa James Tam, “Stereoscopic Image Generation Based on Depth Images for 3D TV,” IEEE Transactions on Broadcasting, June 2005.

    [10] Fehn C, La Barre E., Pastoor S, “Interactive 3-DTV-concepts and key technologies,” Proceedings of the IEEE, Vol. 94, Issue 3, March 2006

    [11] Richard Hartley, “Multiple View Geometry in Computer Vision,” Second Edition, Cambridge University Press, 2003

    [12] Rafael C. Gonzalez and Richard E. Woods, “Digital Image Processing”, Second Edition, Addison Wesley, 2002

    [13] 井上誠喜, C語言數位影像處理, 修訂版, 全華出版社, 2006

    下載圖示 校內：2009-07-18公開
    校外：2010-07-18公開
    QR CODE