本篇論文透過影像圖形本身的灰階值和梯度資訊發展一套高對稱對比 區域對稱偵測流程，分別取出極值點和圖形輪廓，在篩選出特徵點之後，
透過對稱本身具有的梯度資訊作對稱判斷，再整合影像處理方法，自動化 產生輸入影像的軸對稱位置與最大對稱範圍，以達到本篇論文一開始的主 要目的。並透過一連串的實驗與觀察後，可以更加地了解影像圖形對稱特 性，以驗證本系統對於對稱的處理之極限所在。
藉由5.5中的說明，可以了解到此方法在某些類型影像圖形之處理上還 有待加強，因此提出下列幾點可以改進與加強的地方：
(一) 在於圖形輪廓方法上：
從章節 5.5.2 和 5.5.3 中可以發現到，若是在圖形輪廓上能 夠成功分辨主體和背景，並擷取需要的圖形輪廓後再作對稱 偵測，將可以減少干擾並且提高命中率。
(二) 最佳參數：
本篇論文所找出的參數，是藉由一連串的實驗，搭配吾人視 覺輔助，所找出的可能最佳參數，但並非為唯一最佳參數，
所以希望在未來，能以數學運算推演、歸納出最佳解，使本 方法在對稱偵測的判斷上能更加地準確。
- 87 -
在未來工作上，假設能在上述這兩方面都獲得改進，對於偵測準確度 和正確性將能大大獲得提升。
另外，由於本論文所解決的對稱問題主要是軸對稱問題，希望在未來 能透過本方法的延伸或進步來解決更多類型的對稱問題，使其方法能夠更 加地完整、全面。
- 88 -
Reference
[1] J. Wagemans, "Detection of visual symmetries," Spatial Vision, 1995.
[2] H. Weyl, "Symmetry", Princeton University Press, 1952.
[3] Zhitao Xiao and Jun Wu, "Analysis on Image Symmetry Detection Algorithms," FSKD 2007. Fourth International Conference on Fuzzy Systems and Knowledge Discovery, 2007.
[4] Seungkyu Lee and Yanxi Liu, "Curved Glide-Reflection Symmetry Detection," CVPR 2009.
IEEE Conference on Computer Vision and Pattern Recognition, 2009.
[5] Chun-Liang Chiang and Cheng-Chung Lin, "Basic Observations on Potential Cues Behind Vision Target Perception : Intensity, Contrast and Gradient Organization," NCTU Master thesis, 2010.
[6] Ming-Hung Ho and Cheng-Chung Lin, "Symmetry Detection Based on Gradient-Derived Information," NCTU Master thesis, 2011.
[7] Yosi Keller and Yoel Shkolnisky, "An algebraic approach to symmetry detection," ICPR 2004. Proceedings of the 17th International Conference on Pattern Recognition, 2004.
[8] Yuanhao Gong, Chenhui Yang, Qicong Wang and Yahui Gao, "A Novel Symmetry Detection Method for Images and Its application for Motion Deblurring," MMIT '08.
International Conference on MultiMedia and Information Technology, 2008.
[9] L. Lucchese, "A Frequency Domain Algorithm for Detection and Classification of Cyclic and Dihedral Symmetries in Two-Dimensional Patterns," Proceedings. 2002 International Conference on Image Processing, 2002.
[10] Georgios Tzimiropoulos, Vasileios Argyriou and Tania Stathaki, "SYMMETRY DETECTION USING FREQUENCY DOMAIN MOTION ESTIMATION TECHNIQUES," ICASSP 2008. IEEE International Conference on Acoustics, Speech and Signal Processing, 2008.
[11] Daniel Reisfeld, Haim Wolfson and Yehezkel Yeshurun, “Detection of Interest Points Using Symmetry,” Proceedings, Third International Conference on Computer Vision, 1990.
[12] Y. Bonneh, D. Reisfeld and Y. Yeshurun, "Texture Discrimination by Local Generalized Symmetry," Proceedings., Fourth International Conference on Computer Vision, 1993.
[13] Yoram Bonneh, Daniel Reisfeld and Yehezkel Yeshurun, "Quantification of Local
Symmetry: Application for Texture Discrimination," Spatial Vision, vol. 8(4), pp. 515-530, 1994.
[14] Gareth Loy and Jan-Olof Eklundh, "Detecting Symmetry and Symmetric Constellations of Features," Proceeding ECCV'06 Proceedings of the 9th European conference on
Computer Vision, vol. Part II, pp. 508-521, 2006.
- 89 -
[15] Rafael C. Gonzalez and Richard E. Woods, "Digital Image Processing, Third Edition", 2010.
[16] JIUNN-LIN WU, "An Automatic Enhancement Method for High Contrast Images via Bilateral Filtering," Journal of Science and Engineering Technology, vol. 3, no. 3, pp.
81-87, 2007.
[17] S. Paris, P. Kornprobst, J. Tumblin and F. Durand, "Bilateral Filtering: Theory and
Applications," Foundations and Trends in Computer Graphics and Vision, vol. 4, no. 1, pp.
1-73, 2008.
[18] D. G. Low, "Object Recognition from Local Scale-Invariant Features," The Proceedings of the Seventh IEEE International Conference on Computer Vision, 1999.
[19] Hugo Cornelius, Michal Perd’och, Jiri Matas and Gareth Loy, "Efficient Symmetry Detecting Using Local Affine Frames," SCIA 2007, LNCS 4522, pp. 152-161, 2007.
[20] Hugo Cornelius and Gareth Loy, "Detecting Bilateral Symmetry in Perspective,"
Perceptual Organisation in Computer Vision (POCV), 2006.
[21] Milan Sonka, Vaclav Hlavac and Roger Boyle, "Image Processing, Analysis, and Machine Vision".