在東方國家，印章在金融、行政方面被廣泛地用於驗證個人或組織的身份，
因此有效地識別印章真偽的能力是不可或缺的。

本文針對方形章及圓形章分別提出不同的參考點定位方法，
並在定位後進行幾何轉換以實現圖像配准，
修正圖像取得過程中可能存在的視角不一致所造成的幾何畸變。

藉由分析印章圖像組內差異的來源，我們提出了一個合理的假設，
並定義數種相似度以量化印章圖像間之差異。

取得參考圖像及待測圖像間的邊緣差異後，
我們藉由距離變換產生我們所定義的數種加權矩陣，
最終利用加權矩陣及邊緣差異圖形計算相似度，
並將其作為支援向量機的輸入向量，以判斷印章真偽。

實驗結果顯示不論是以用戶依存或非用戶依存的訓練方法，
我們所提出方法都具有一定的辨識能力。

In Asian countries, seals are widely used for authenticating the identity of a person or organization. Therefore, the ability to efficiently verify whether a seal is either genuine or forged is important.

We use different method to approximate pairs of reference points in reference image and detected imprint image for rectangular and circular imprint, respectively.

Then, geometric transformation is performed to align
the perspective of the detected imprint image with the reference
imprint.
After the edge-difference images between the reference image
and the detected image are created, distance transformation is applied to get several kinds of weighting matrices.

Finally, the edge-difference images are weighted element-wisely with the edge-difference images to calculate some predefined similarities which are used as the input vector of the SVM (support vector machine).

The imprint is then determined to be either genuine or forged. The experimental results show the effectiveness of the proposed verification approach.

Abstract I
中文摘要II
1 Introduction 1
2 Preliminaries 5
2.1 Imprint Extractions . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2 Image Registration . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.3 FEATURE EXTRACTION . . . . . . . . . . . . . . . . . . . . 10
2.4 SUPPORT VECTOR MACHINE . . . . . . . . . . . . . . . . . 15
3 PROPOSED SYSTEM 17
3.1 Imprint extraction . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2 Shape recognition . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.3 Reference points locating . . . . . . . . . . . . . . . . . . . . . . 20
3.4 Geometric transformation . . . . . . . . . . . . . . . . . . . . . 21
3.5 SVM-based VERIFICATION . . . . . . . . . . . . . . . . . . . 24
3.5.1 Proposed Hypotheses . . . . . . . . . . . . . . . . . . . . 24
3.5.2 Proposed similarities . . . . . . . . . . . . . . . . . . . . 26
4 EXPERIMENT RESULTS 34
5 Conclusion 43
Bibliography 44

[1] J. Serra, "Biomedical image analysis by mathematical morphology (au-thor's transl)," Pathologie-biologie, vol. 27, no. 4, pp. 205-207, 1979.
[2] T.-J. Fan and W.-H. Tsai, \Automatic chinese seal identification," Com-puter vision, graphics, and image processing, vol. 25, no. 3, pp. 311-330,1984.
[3] W. Pan, J. Hu, M. Liu, X. Qi, and H. Lang, "Seal imprint segmentation
based on color feature classifier," in 2012 International Conference on
Audio, Language and Image Processing. IEEE, 2012, pp. 837-840.
[4] K. Ueda, "Extraction of signature and seal imprint from bankchecks by
using color information," in Proceedings of 3rd International Conference on Document Analysis and Recognition, vol. 2. IEEE, 1995, pp. 665-668.
[5] H. Lang, C. Xie, X. Qi, and H. Ling, "Seal forgery detection by geometric
consistency," Information: An International Interdisciplinary of Journal,2012.
[6] K. Ueda and K. Matsuo, "Automatic seal imprint verication system for
bank check processing," in Third International Conference on Information Technology and Applications (ICITA'05), vol. 1. IEEE, 2005, pp. 768-771.

[7] X.Wang and Y. Chen, "Seal image registration based on shape and layout characteristics," in 2009 2nd International Congress on Image and Signal
Processing. IEEE, 2009, pp. 1-5.
[8] T. Horiuchi, "Automatic seal verification by evaluating positive cost," in
Proceedings of Sixth International Conference on Document Analysis and
Recognition. IEEE, 2001, pp. 572-576.
[9] H. Zhang and J. He, "Automatic seal imprint verification by quanti-fying edge difference," in Optoelectronic Imaging and Multimedia Tech-nology, vol. 7850. International Society for Optics and Photonics, 2010,p. 78500D.
[10] J. He, H. Zhang, and T. Liu, "Seal imprint verification using edge difference histogram," in Optoelectronic Imaging and Multimedia Technology
II, vol. 8558. International Society for Optics and Photonics, 2012, p.855804.
[11] S. Hua, "Researches on graph matching based character recognition for
chinese seal images." Master's Thesis of Department of Control engi-
neering . Hunan: Hunan University, 2018.
[12] W. Liu, "Researches on graph matching based character recognition for
chinese seal images." Master's Thesis of Department of Signal and in-formation processing. Shandong: Qingdao University, 2018.

[13] R. O. Duda and P. E. Hart, "Use of the hough transformation to detect
lines and curves in pictures," SRI INTERNATIONAL MENLO PARK
CA ARTIFICIAL INTELLIGENCE CENTER, Tech. Rep., 1971.
[14] C. Cortes and V. Vapnik, "Support-vector networks," Machine learning,vol. 20, no. 3, pp. 273{297, 1995.
[15] R. C. Gonzalez, R. E. Woods et al., "Digital image processing [m]," Pub-lishing house of electronics industry, vol. 141, no. 7, 2002.
[16] R. Haruki, T. Horiuchi, H. Yamada, and K. Yamamoto, "Automatic seal
verification using three-dimensional reference seals," in Proceedings of
13th International Conference on Pattern Recognition, vol. 3. IEEE,
1996, pp. 199-203.
[17] R. Fabbri, L. D. F. Costa, J. C. Torelli, and O. M. Bruno, "2d euclidean distance transform algorithms: A comparative survey," ACM Computing Surveys (CSUR), vol. 40, no. 1, p. 2, 2008.
[18]L. Lam, S. . Lee, and C. Y. Suen, "Thinning methodologies-a compre-hensive survey," IEEE Transactions on Pattern Analysis and Machine
Intelligence, vol. 14, no. 9, pp. 869-885, Sep. 1992.
[19] J. Serra, "Introduction to mathematical morphology," Computer vision,
graphics, and image processing, vol. 35, no. 3, pp. 283-305, 1986.
[20] X. Wang and Y. Chen, "A novel seal imprint verification method based
on analysis of difference images and symbolic representation," in International Workshop on Computational Forensics. Springer, 2010, pp. 56-67.
[21] G. S. Eskander, R. Sabourin, and E. Granger, "Hybrid writer-
independent{writer-dependent offline signature verification system," IET biometrics, vol. 2, no. 4, pp. 169-181, 2013.