تشخیص برون‌خط جعل دست خط فارسی غیر وابسته به متن با استفاده از تحلیل بافت

نوع مقاله : مقاله پژوهشی

نویسندگان

1 شهید بهشتی

2 شهید بهشتی تهران

چکیده

امروزه دست­خط به عنوان یک زیست‌سنج رفتاری شناخته شده و مورد قبول است. همراه با گسترش این مقبولیت سوء استفاده ازآن نیز گسترش یافته است. در این مقاله روشی برای تشخیص برون‌خط جعل دست‌خط فارسی در حالت غیر وابسته به متن ارائه شده است. این روش مبتنی بر استخراج دو بافت متراکم از حروف پایه و جزییات نوشته می‌‌باشد. پس از دودویی کردن و استخراج اسکلت نوشته و اعمال متوالی فیلتر گابور و تدریج فاز محلی بردار ویژگی نمونه‌ها استخراج و در نهایت با استفاده از دسته‌بندی یک کلاسه و تنها با وجود نمونه‌های مثبت داده‌ها دسته‌بندی شده وجعلی بودن یا نبودن آنها بررسی می‌شود. استفاده از ترکیب این دو توصیف‌گر برای دادگان فارسی  برای اولین بار در این مقاله به کار گرفته شده است. از آنجا که هیچ مجموعه داده مناسبی در زبان فارسی برای این منظور وجود نداشت، یک مجموعه داده از 62 نویسنده با دو نمونه نوشته جمع­آوری شد. در نمونه اول از نویسندگان درخواست شد که متنی را بازنویسی کنند و در نمونه دوم از آنها خواسته شد درباره یک تصویر مطلبی بنویسند. برای تولید نمونه­های جعلی از هنرجویان ممتاز خط خواسته شد که نمونه دوم را تقلید کنند تا برای هر نویسنده سه مورد نمونه جعل تولید شود. نتایج آزمایش‌ها رضایت‌بخش بوده و با دقت بیش از 84 درصد کیفیت روش پیشنهادی را نشان می‌دهد.

کلیدواژه‌ها


عنوان مقاله [English]

Offline text-Independent Persian handwriting Forgery Detection using Texture Analysis

نویسندگان [English]

  • E. Ghanbari Maman 1
  • M. Ebrahimi Moghaddam 2
1
2
چکیده [English]

Nowadays handwriting is known and accepted as a behavioral biometric. Handwriting-based             authentication has become a hot topic of research as a result of increasing handwriting forgery. In this   research an offline text-independent method for detection of forgeries in Persian handwritten documents is presented. The proposed approach is based on extracting two dense textures of basic alphabet and some text details. A one class classification (OCC) with only positive (original) samples is used in classification together with Binarization and Skeletonization in the preprocessing phase. Also, a combination of Gabor filter and LPQ descriptor is used to complete the feature vector. Both descriptors have been used for writer identification and verification in English and Persian but this is the first time that the combination of these two is used on a Persian handwriting. We also introduce a new dataset from 62 individuals with 2 pages per each. In the first instance the writers were instructed to copy a preset text and in the second one they were asked to write at least 6 lines about a picture. These instances were to be used as training and testing samples respectively. In order to generate counterfeit samples, we asked 5 qualified calligraphers to       duplicate the second sample of each writer. Three forgeries per writer were generated in this way. Tests results are satisfactory and present more than 84% accuracy, demonstrating the quality of the proposed method.
 

کلیدواژه‌ها [English]

  • Biometrics
  • forgery detection
  • texture analysis
  • text-independent
[1]     H. Arora, S. Lee, S. Srihari, and S. Cha, “Individuality of Handwriting,” Journal of Forensic Sciences, vol. 47, pp. 1-17, 2002.##
[2]     M. Sreeraj and S. Idicula, “A survey on writer identification schemes, International Journal of Computer Application, vol. 26, pp. 23-33, 2011.##
[3]     M. Bulacu and L. Schomaker, “Text-Independent Writer Identification and Verification Using Textural and Allographic Features”,  IEEE transactions on pattern analysis and machine intelligence, vol. 29, 2007.##
[4]     B. Helli  and M. E. Moghaddam, “An off-line cheque handwritten forgery detection based on feature route density matrix”, Pattern Analysis and Applications, vol. 17, 2014.##
[5]     S. Cha, C. Tappert, M. Gibbons, and Y. Chee, “Automatic Detection of Handwriting Forgery using a Fractal Number Estimate of Wrinkliness”,  International Journal of Pattern Recognition and Artificial Intelligence, vol. 18, pp.        1361-1371, 2004.##
[6]     D. Bertolini, L. S. Oliveira, E. Justino, and R. Sabourin, “Assessing textural features for writer identification on different writing styles and forgeries”,  in 22nd International Conference on Pattern Recognition (ICPR), 2014.##
[7]     H. Said, T. Tan, and K. Baker, “Personal identification based on handwriting”, Pattern Recognition, vol. 33, 2000.##
[8]     F. Shahabinejad  and M. Rahmati, “A New Method for Writer Identification and Verification Based on Farsi/Arabic Handwritten Texts”, in Ninth International Conference on Document Analysis and Recognition, 2007.##
[9]     M. Lutf, X. You, and H. Li, “Offline Arabic handwriting identification using language diacritics”, in 20th International Conference on Pattern Recognition (ICPR), 2010.##
[10]  C. Fahn, C. Lee, and H. Chen, “A text independent handwriting forgery detection system based on branchlet features and Gaussian mixture models”,  in  14th Annual Conference on Privacy, Security and Trust (PST), 2016.##
[11]  B. Helli and M. E. Moghaddam, “A text-independent Persian writer identification based on feature relation graph (FRG)”,  Pattern Recognition, pp. 2199-2209, 2010.##
[12]  B. Helli and M. E. Moghaddam, “A writer identification method based on XGabor and LCS,” IEICE Electronics Express,  vol. 6, pp. 623-629, 2009.##
[13]  H. Yaâcoub, I. Siddiqi, and M. Kettani, “Arabic Writer Identification Using Local Binary Patterns (LBP) of Handwritten Fragments”, Pattern Recognition and Image Analysis, 2015.##
[14]  X. Wu, Y. Tang, and W. Bu, “Offline Text-Independent Writer Identification Based on Scale Invariant Feature Transform”, IEEE transactions on information forensics and security, vol. 9, 2014.##
[15]  A. Nicolaou, A. Bagdanov, M. Liwicki, and D. Karatzas, “Sparse Radial Sampling LBP for Writer Identification”, arXiv preprint arXiv, vol. 1504, 2015.##
[16]  M. Y. A. A. A. a. K. K. Khan, “Deep learning for automated forgery detection in hyperspectral document images”, Journal of Electronic Imaging, vol. 5, 2018.##
[17]  V. Ojansivu and J. Heikkilä, “Blur insensitive texture classification using local phase quantization”, Image and signal processing, 2008.##
[18]  T. Ojala, M. Pietikäinen, and D. Harwood, “A Comparative Study of Texture Measures with Classification Based on Feature Distributions”, Pattern Recognition, vol. 29, pp.   51-59, 1996.##
[19]  Z. Guo, L. Zhang, and D. Zhang, “A completed modeling of local binary pattern operator for texture classification,” IEEE Transactions on Image Processing,  vol. 19, 2010.##
[20]  X. Tan and B. Triggs, “Enhanced local texture feature sets for face recognition under difficult lighting conditions,” IEEE Transactions on Image Processing, pp. 1635-1650, 2010.##
[21]  T. Lee, “Image representation using 2D Gabor wavelets”, IEEE Transactions onPattern Analysis and Machine Intelligence,  vol. 18, pp. 959-971, 1996.## 
[22]  H. Chen, “Robust text detection in natural images with  edge-enhanced Maximally Stable Extremal Regions”, in 18th IEEE International Conference on Image Processing (ICIP), Brussels, 2011.##
[23]  D. Marr and E. Hildreth, “Theory of edge detection”, Proceedings of the Royal Society of London B: Biological Sciences, pp. 187-217, 1980.##
[24]  L. Reginald, “Basic Methods for Image Restoration and Identification, ”Information and Communication Theory Group, Delft University of Technology, Elsevier, 1997.##
[25]  D. Bertolini, L. Oliveira, E. Justino, and R. Sabourin, “Texture-based descriptors for writer identification and verification”, Expert Systems with Applications,  vol. 40, pp. 2069-2080, 2013.##
[26]  J. Bhatt and N. Patel, “a survey one class classification using ensembles method”, International Journal for Innovative Research in Science and Technology, vol. 7, 2015.##
[27]  R. Safabaksh, A. Ghanbarian, and G. Ghiasi, “HaFT: A handwritten Farsi text database”,  in Iranian Conference on Machine Vision and Image Processing (MVIP), 2013.##
[28]  M. Ziaratban, K. Faez, and F. Bagheri, “FHT: An Unconstraint Farsi Handwritten Text Database”,  in International Conference on Document Analysis and Recognition (ICDAR), 2009.##
[29]  H. El Abed and V. Margner, “The IFN/ENIT-database - a tool to develop Arabic handwriting recognition systems”,  in  International Symposium on Signal Processing and Its Applications. ISSPA, 2007.##
[30]  V. Pervouchine and G. Leedham, “Extraction and analysis of forensic document examiner features used for writer identification”,  Pattern Recognition,  vol. 40, pp. 1004-1013.##
[31]  M. Malik, M. Liwicki, L. Alewijnse, W. Ohyama, M. Blumenstein , and B. Found, “Signature verification and writer identification competitions for on-and offline skilled forgeries (sigwicomp)”,  in 12th Int. Conf. on Document Analysis and Recognition, 2013.##