臺灣博碩士論文加值系統

近年來由於手機相機的便利性，造成人人隨手都可以拍攝，但傳統照像所能拍攝的影像範圍有限，在需要大範圍的視野要進行拍照需使用拍攝多張後再行拼圖，故在本篇論文中，以魚眼鏡頭進行拍攝，得到原始魚眼拍攝後扭曲的影像，進行影像扭曲校正，讓扭曲的影像復原為一般廣角的影像。
魚眼鏡頭所拍攝的影像通常會造成嚴重畸形，一般以影像中心為圓點往四周擴散，造成四周變為一個圓狀物，而本論文將四周扭曲的部份使用張友正所提出的影像扭曲校正的方式來達到四周影像扭曲的校正並使用三種不同的魚眼鏡頭角度各自使用本論文校正方式進行校正，最後進行三種不同魚眼鏡頭影像校正後的比對。
本論文經由不同角度的魚眼鏡頭進行校正，除了可以校正小角度的魚眼鏡頭影像亦可校正大於水平面角度的魚眼鏡頭，顯示本實驗結果可以搭配不同的魚眼鏡頭進行相關的影像扭曲校正。

In recent years, everyone can take photos everywhere when he/she wants because of the popularity and convenience of the mobile cameras. However, common cameras have limited image scope. If we want to take wide-angle image photos, we need to take several photos and then splice. In this thesis, we discuss taking original distorted image photos by using curved lens and fisheye lens, and then proceeding image distortion correction to get common wide-angle image photos.
The image photos taking by curved lens and fisheye lens usually would be seriously distorted – in common cases, spreading toward the edges from the image center leading to rounded edge images. In this thesis, we use the image distortion correction method disclosed by Chang to conduct the distortion correction of the rounded edge images, and use three different kinds of curved lens angles to have distorted image photos first and then proceed the distortion correction to compare the results.
By comparing the results of the distortion correction of the image photos taking by three different kinds of curved lens and fisheye lens, we found that not only the small angle curved lens images can be distortion corrected but also the wide angle fisheye lens images. It shows that we can conduct the image distortion correction of the photos taking by different angle curved lens or fisheye lens.

摘要 i
Abstract ii
目錄 iii
表目錄 v
圖目錄 vi
第1章 緒論 1
1.1 前言 1
1.2 研究動機與目的 1
1.3 問題與陳述 2
1.4 方法與貢獻 2
1.5 論文架構 3
第2章 文獻探討 4
2.1 魚眼鏡頭 4
2.1.1 鏡頭畸變 5
2.2 相機校正 9
2.2.1 相機座標轉換 10
2.2.2 影像校正的空間點轉換 14
2.3 針孔鏡頭成像 17
2.4 魚眼鏡頭成像模型 18
2.4.1 2D球面坐標定位法 19
2.4.2 多項式坐標變換算法 20
2.4.3 球面透視投影算法 21
第3章 研究方法 23
3.1 系統架構 23
3.1.1 系統流程與相關理論 26
3.1.2 影像校正棋盤格 27
3.1.3 影像灰階化 28
3.1.4 影像二值化 29
3.1.5 角點與亞像素點擷取 33
3.1.6 相機內外參數 35
第4章 系統實現與實驗結果 38
4.1 實驗環境 38
4.1.1 系統實驗環境規格 38
4.1.2 系統實驗硬體規格 38
4.1.3 本論文實驗的程式介面 40
4.2 實驗系統流程說明 41
4.2.1 系統方塊圖 41
4.2.2 棋盤格位置圖 42
4.2.3 棋盤格影像灰階化 44
4.2.4 棋盤格影像灰階化後的二值化 45
4.2.5 取角點與亞像素(Sub-Pixel) 46
4.2.6 取得內外參數平均誤差 47
4.3 1張影像、2張影像、3張影像與10張影像校正後比對 61
4.4 各鏡頭扭曲前後的相關比對 62
第5章 結論與未來展望 65
5.1 結論 65
5.2 未來展望 66
參考文獻 67

[1]S. Y. Wei,” Design and Implementation of Real-time Continuous Image Mosaic System,” Master Thesis, Department of Computer Science, National Central University ,2010. (in Chinese)
[2]S. B. Zhao, B. F. Zhang, L. Li, Z. L. Cao.,” Fisheyes Lens Camera System Calibration and Localization Error Analysis,” 2010 International Conference on Computer Design And Applications (ICCDA 2010).
[3]H. Yuan.,” The Research and Implementation of Algorithm for Real-time Fisheye Lens Video Image Calibration,” 2007. Master Thesis, Dept. Computer Science and Information Engineering., Sun Yat-sen Univ. 2007.
[4]T. N. Mundhenk, J.Michae1, X.Q. Liao, and E. L. Hall,” Techniques for Fisheye Lens Calibration using a Minimal Number of Measurements” Procof the SPIE Intelligent Robotics and Computer Vision Conference, Boston, Massachusetts，2000，v01．11，pp：8—9．.
[5]G. Wolberg,” Digital Image Warping,” IEEE Computer Society Press, Los Alamitos, CA, USA, 1994.
[6]F. Devernay, O. Faugeras,” Straight Lines Have to Be Straight,” IN SPIE VOLUME2567, vol. 2567,2001
[7]J. S. Sheu,” Implementing a 3D Vision Robot System with Integrated Position Tracking and Object Distance Calculation,” Intelligent Engineering Systems through Artificial Neural Networks, (IESANN), (ISSN 2220-5306), Vol. 21, Issue 2, pp. 175–188, March 2011.
[8]D.C. Brown.,” Close-range camera calibration.” Photogrammetric Engineering, 37(8):855-866,1971
[9]G.P. Stein,” Lens distortion calibration using point correspondences” Computer Vision and pattern Recognition, 1977. Proceedings., 1977 IEEE Computer Society Conference on , vol., no., pp.602-608, 17-19 Jun 1977.
[10]C. H. Hsu,” A Distortion Correction Method for Endoscopy Images Based on Calibration Patterns and Parameters of Optical Lens,” Master Thesis, Department of Electronic Engineering, Chung Yuan Christian University ,2004. (in Chinese)
[11]J. J.More.,” The Liebenberg-Marquardt algorithm : Implementation and theory” Conference on Numerical Analysis University of Dundee Soot land June28-July1, 1977
[12]J. I. Kunovský,” Modern Taylor Series Method” Scientific Conference on Informatics, 2015 IEEE 13th International
[13]K. Wang, L. J. Zhao, R. F. Li.,” Fisheye omnidirectional camera calibration –Pinhole or spherical model？” Conference on Robotics and Biomimetics (ROBIO), 2014 IEEE International
[14]X. Li, W. Liu, W. Fan, J. Sun, Naoi Satoshi.,” Perspective correction using camera intrinsic parameters” Conference on Signal Processing (ICSP), 2016 IEEE 13th International
[15]X. F. Wang, W. J. Feng, Q. J. Liu, B. F. Zhang, Z. L. Cao.,” Calibration Research on Fish-eye lens” Conference on information and Automation(ICIA), 2010 IEEE International
[16]K. Miyamoto.,” Fish Eye Lens” Journal of the Optical Society of America , vol.54,Issue 8, pp. 1060-1061(1964).
[17]F. Battonvil,” Fisheye lenses” Journal of the International Meteor Organization, vol. 33,no. 1,p.9-14 (2005)
[18]J.H Park, B.U. Lee, S.C. Byun.,” Lens Distortion Correction Using Ideal Image Coordinates” IEEE Transactions on Consumer Electronics, vol. 55,Issue 3,August 2009
[19]R. Hartley, S.B. Kang.,” Parameter-Free Radial Distortion Correction with Center of Distortion Estimation” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 29, No. 8, August 2007
[20]T. Maruyama, I. Niitsu, K. Osakabe,” Pincushion Distortion-Free Self-Convergence Deflection System” IEEE Transactions on Consumer Electronics, vol. CE-25, Issue 4,Aug.1979
[21]X. Z Wang, X. Z. Bu, B. Yang, G. J. Li, J. Yu.,” Corrected unwrapping method based on the tangential and radial distortion center for the panoramic annular image” IEEE IET Image Processing , vol. 9, Issue 2, Feb.2015
[22]D. Scaramuzza, A. Martinelli, R. Siegwart.,” A Flexible Technique for Accurate Omnidirectional Camera Calibration and Structure from Motion” Proceedings of the Fourth IEEE International Conference on Computer Vision Systems (ICVS 2006)
[23]Z. Zhang.,” A Flexible new technique for camera calibration” IEEE Transactions on Pattern Analysis and Machine Intelligence (Volume:22, Issue:11, Nov 2000)
[24]V. Popescu, B. Benes, P. Rosen, J. Cui, L.L. Wang.,” A Flexible Pinhole Camera Model for Coherent Nonuniform Sampling” IEEE Computer Graphics and Applications (Volume:34, Issue:4 July-Aug.2014)
[25]K. Wang, L. J. Zhao, R. F. Li.,” Fisheye omnidirectional camera calibration- Pinhole or spherical model？” Robotics and Biomimetics (ROBIO), 2014 IEEE international Conference on
[26]Y.T. Do.,” Optical Aberration Correction by Divide-and-Learn for Accurate Camera Calibration” IEEE Computational Science and Computational Intelligence (CSCI), 2014 International Conference on
[27]W.W. Kim, S.H. Park, J.K. Hwang, S.Y. Lee.,” Automatic head pose estimation from a single camera using projective geometry” IEEE Information, Communications and Signal Processing (ICICS) 2011 8th International Conference on
[28]W. N. Xi, L. Jun, J. P. Yan,” An Improved Intrinsic Parameter Calibration Method” Master Thesis, Department of Mechanical Engineering of Southeast University. Nanjing ,2001. (in Chinese)
[29]A. Gavriilidis, C.S Schmidt, J. Velten, A. Kummert,” Evaluation of the Lens Distortion Errors in 2-D Image to 3-D World Coordinate Projections” IEEE Multidimensional Systems (nDS), 2013. Proceedings of the 8th International Workshop on
[30]H. Farid, A.C. Popescu,” Blind Removal of Lens Distortion” Journal of the Optical Society of America A,2001.
[31]X. L. Wang, M. Li, Y. X. Cui.,” An improved method of nonlinear distortion correction for wide-angle lens” 2016 IEEE International Conference on Signal and Image Processing (ICSIP), Pages:69-73
[32]R. Q. Wu, L. L. Wang, X. R. Cheng.,” A piecewise correction method for high-resolution image of wide-angle lens” IEEE 2013 6th International Congress on Image and Signal Processing (CISP), Volume:01, Pages:331-335
[33]N. Hottong, S. Becker, S. Beuck.,” Evaluation of observers’ sensitivity to lens distortion discrepancies in stereoscopic moving pictures” IEEE 2016 January 28, 3D Imaging(IC3D), 2015 International Conference on.
[34]J. Zhang, Z. Z. Wang, Z. L. Yang ,”Correction of single circular fisheye image” ISSN 1001-9081 Journal of Computer Application 2015,35(5):1444-1448
[35]S. Shishir, J.K. Aggarwal.,”Intrinsic parameter calibration procedure for a (high-distortion) fish-eye lens camera with distortion model and accuracy estimation” Pattern Recognition 29.11 (1996): 1775-1788.
[36]B. K. Zhang, J. K. Li, J. Wang, X. Y. Wang.,” Fisheye Lens Distortion Correction Based On An Ellipsoidal Function Model” 2015, International Conference on Industrial Informatics-Computing Technology, Intelligent Technology, Industrial Information Integration.
[37]C. C. Kang,” Mobile Robot Automatic Parking with Stereo Vision Range Estimation,” Master Thesis, Department of Mechanical Engineering, National Taiwan University of Science and Technolog,2011. (in Chinese)
[38]Y. C. Liu,” Fisheye Camera Calibration and Multiple-Viewpoint Image Stitching,” Master Thesis, Department of Computer Science, National Chiao Tung University,2007. (in Chinese)
[39]B. F. Zhang, Z. Q. Qi, J. C. Zho, Z. L. Cao.,” Omnidirectional image restoration based on spherical perspective projection” Circuits and Systems, 2008. APCCAS 2008. IEEE Asia Pacific Conference on
[40]Y. L. Chen, J. S. Sheu, C. M. Cheng and W. H. Tsai,” Implementation of Robot Platform in Face Detection and Tracking Based on a New Authentication Scheme,” Advances in Mechanical Engineering, Vol. 2014, pp. 1-9, 2014.