離線手寫簽名驗證是掃描個人簽名影像，經過計算機視覺處理和自定義特徵擷取函數擷取特徵，再將特徵分析後完成手寫簽名驗證。儘管現在有良好的特徵擷取函數以及分類器能分出真正和偽造的簽名，但對一個高內變異性的資料如個人簽名，則很難獲得到關鍵的特徵，驗證效果就不理想。

故我們使用深度孿生卷積網路 (Siamese Network) 來學習特徵，並且訓練出個別的特徵擷取模型。與一般的卷積神經網路不同在於它訓練時是一次丟兩張圖片下去訓練，最後的輸出是一個相似值而非分類真偽。在這篇論文的離線手寫簽名驗證系統中，有實作 Writer-Dependent (WD) 系統和 Hybrid 系統兩種模式。WD 系統是指為每個使用者建立個別的模型(Personal Model)來訓練特徵並決定出真偽，驗證時再用自己的模型來獲取特徵和決定真偽；Hybrid 系統則是結合 Writer-Independent (WI) 和 WD 的方法，首先將一部分的使用者放進一個模型訓練特徵(WI Feature Learning)，再為其他的使用者對其建立個別的分類器來識別真偽(WD Classification)，使其成為新的系統架構。

實驗採用 GPDS 和 ICDAR 2011 兩個離線簽名資料庫，在 ICDAR 2011 表現最好的準確率為 96.88%；在GPDS 表現最好的準確率為 92.3%。

Off-line handwritten signature verification is to scan the personal signature image, and after computer visual processing and custom feature extraction function capture feature, the feature analysis is completed to the handwritten signature verification. Although there are good feature extraction functions and classifiers that can separate genuine and forged signatures, it is difficult to obtain key features for a high intra-variability data such as a personal signature, and the verification effect is not ideal.
Therefore, we use the Deep Siamese Convolutional Neural Networks to learn the features. It's different from the general network training method. When you input two pictures at a time, the final output is a similar value, not the classification. In the off-line handwritten signature verification system of this paper, there are two modes: Writer Dependent (WD) system and Hybrid system. The WD system refers to create of a personal model for each user to train features and verify identity by the personal model. The Hybrid system is a part of the user who put in a feature learning model(WI) and creates an individual classifier for other users to identify (WD).
The experiment uses two offline signature databases: GPDS and ICDAR 2011. The best accuracy in ICDAR 2011 is 96.88%. In GPDS, the best accuracy is 92.3%.

摘要
目錄
一、介紹----------------------------------------1
二、手寫簽名技術回顧--------------------------7
三、離線混合型簽名驗證系統-------------------16
四、離線作者相依簽名驗證系統-----------------34
五、結論與未來展望-----------------------------39
參考文獻----------------------------------------40

1. L. G. Hafemann, R. Sabourin, and L. S. Oliveira, "Offline Handwritten Signature Verification-Literature Review", Computer Vision and Pattern Recognition(CVPR), arXiv preprint arXiv:1507.07909 , 2015.
2. Ashok Kumar and Karamjit Bhatia, "A survey on offline handwritten signature verification system using writer dependent and independent approaches", 2016 2nd International Conference on Advances in Computing, Communication, and Automation (ICACCA) (Fall), International Conference on. IEEE, 2016, pp. 1–6.
3. M. K. Kalera, S. N. Srihari, A. Xu. Offline signature verification and identification using distance statistics. IJPRAI(International Journal of Pattern Recognition and Artificial Intelligence) 18 (7) (2004) 1339–1360.
4. G. Dimauro, S. Impedovo, G. Pirlo, A. Salzo, "A multi-expert signature verification system for bankcheck processing", IJPRAI 11 (05) (1997) 827–844.
5. Y. Serdouk, H. Nemmour, and Y. Chibani. Off-line handwritten signature verification using variants of local binary patterns. Networking and Advanced Systems, 2nd International Conference on, p. 75, 2015.
6. Baltzakisa H., Papamarkosb N. A new Signature Verification Technique based on a Two Stage Neural Network Classifier. Elsevier Journal Engineering Applications of Artificial Intelligence 14, pp. 95 – 103, 2001.
7. Y. Lecun, L. Bottou, Y. Bengio and P. Haffner, "Gradient-based learning applied to document recognition". Proceedings of the IEEE 86, 11, 2278 – 2324, Nov. 1998.
8. Alex Krizhevsky, Ilya Sutskever, and Geoffrey E. Hinton, "ImageNet Classification with Deep Convolutional Neural Networks". In F. Pereira, C. J. C. Burges, L. Bottou, and K. Q. Weinberger, eds. Advances in Neural Information Processing Systems 25 (NIPS), 2012.
9. Karen Simonyan and Andrew Zisserman, "Very Deep Convolutional Networks for Large-Scale Image Recognition". Computer Vision and Pattern Recognition, 2014.
10. Gabe Alvarez, Blue Sheffer, and Morgan Bryant, "Offline Signature Verification with Convolutional Neural Networks". Technical Report. Tech. rep., Stanford University, 2016.
11. Luiz G. Hafemann, Robert Sabourin and Luiz S. Oliveira, "Writer-independent Feature Learning for Offline Signature Verification using Deep Convolutional Neural Networks", IEEE, International Joint Conference on Neural Networks (IJCNN), 2016.
12. Jane Bromley, Isabelle Guyon, Yann LeCun, Eduard Säckinger, and Roopak Shah, "Signature Verification using a Siamese Time Delay Neural Network". Int. J. Pattern Recognit. Artificial Intell., vol. 7, no. 4, pp. 669687, Aug. 1994.
13. F. Schro, D. Kalenichenko, J. Philbin. Facenet: A unified embedding for face recognition and clustering. in: CVPR, 2015, pp. 815–823.
14. S. Chopra, R. Hadsell, Y. LeCun. Learning a similarity metric discriminatively with application to face verification. in: CVPR, 2005, pp. 539–546.
15. Sounak Dey, Anjan Dutta, J. Ignacio Toledo, Suman K. Ghosh, Josep Llados and Umapada Pal, "SigNet: Convolutional Siamese Network for Writer Independent Offline Signature Verification". CVPR(Computer Vision and Pattern Recognition), 2017.
16. Emre Ozgunduz, Tulin Senturk, and M. Elif Karsligil, "Off-line signature verification and recognition by support vector machine". In European signal processing conference, EUSIPCO, 2005.
17. Ashwini Pansare and Shalini Bhatia, "Handwritten Signature Verification using Neural Network". International Journal of Applied Information Systems, Volume 1– No.2, pp.44 – 49, January 2012.
18. Marcus Liwicki, Muhammad Imran Malik, C. Elisa van den Heuvel, Xiaohong Chen, Charles Berger, Reinoud Stoel, Michael Blumenstein and Bryan Found, "Signature Verification Competition for Online and Offline Skilled Forgeries (SigComp2011)". International Conference on Document Analysis and Recognition, IEEE, 2011.
19. M. A. Ferrer, M. Diaz-Cabrera, A. Morales, "Synthetic off-line signature image generation", in: ICB, 2013, pp. 1–7.
20. Digital Signal Processing Group (GPDSsyntheticSignature database) at: http://www.gpds.ulpgc.es/, June 2013.
21. G. Eskander, R. Sabourin, and E. Granger. Hybrid writer-independentwriter-dependent offline signature verification system. IET Biometrics, vol. 2, no. 4, pp. 169–181, Dec. 2013.
22. M. R. Pourshahabi, M. H. Sigari, and H. R. Pourreza. "Offline handwritten signature identification and verification using contourlet transform". In Soft Computing and Pattern Recognition, International Conference of. IEEE, 2009, pp. 670–673.
23. N. Otsu, "A threshold selection method from gray-level histograms". Automatica, vol. 11, no. 285-296, pp. 23–27, 1975.
24. Harris, C. and Stephens, M. ,“A combined corner and edge detector.” In Alvey Vision Conference, pp. 147–151. 1988.
25. Xavier Glorot and Yoshua Bengio, "Understanding the difficulty of training deep feedforward neural networks", Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, PMLR 9:249-256, 2010.
26. N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever and R. Salakhutdinov, "Dropout: A Simple Way to Prevent Neural Networks from Overfitting". The Journal of Machine Learning Research, pages 1929 – 1958, 2014.
27. S. Ioffe and C. Szegedy, "Batch normalization: Accelerating deep network training by reducing internal covariate shift." In ICML, 2015. 3
28. Diederik P. Kingma and Jimmy Ba, "Adam: A Method for Stochastic Optimization", published as a conference paper at the 3rd International Conference for Learning Representations, San Diego, 2015.
29. Lutz Prechelt. "Early stopping-but when?". Neural Networks: Tricks of the trade, pp. 553–553, 1998.
30. Shahab Tayeb et al., "Toward data quality analytics in signature verification using a convolutional neural network". IEEE, International Conference on Big Data, Published 2017.
31. J.F. Vargas, C.M. Travieso, J.B. Alonso, and M.A. Ferrer, "Off-line Signature Verification Based on Gray Level Information Using Wavelet Transform and Texture Features." In International Conference on Frontiers in Handwriting Recognition, pages 587–592, 2010.
32. Mustafa Berkay Yilmaz, Berrin Yanikoglu, Caglar Tirkaz, and Alisher Kholmatov. Offline signature verification using classifier combination of HOG and LBP features. In Biometrics (IJCB), 2011 International Joint Conference on, pages 1–7. IEEE, 2011.
33. B Shekar, R Bharathi, J. Kittler, Yu.V. Vizilter, and L Mestestskiy, "Grid structured morphological pattern spectrum for off-line signature verification." In International Conference on Biometrics, 2015.