بازجویی قانونی چاپگر مبتنی بر بردار هویت حاصل از ناحیه بندی بافت تصویر
محورهای موضوعی : انرژی های تجدیدپذیر
روزبه حمزه ئیان
1
(دانشکده مهندسی برق و کامپیوتر- واحد علوم و تحقیقات، دانشگاه آزاد اسلامی، تهران، ایران)
فربد رزازی
2
(دانشکده مهندسی برق و کامپیوتر- واحد علوم و تحقیقات، دانشگاه آزاد اسلامی، تهران، ایران)
علیرضا بهراد
3
(دانشکده فنی و مهندسی- دانشگاه شاهد، تهران، ایران)
کلید واژه: بازجویی قانونی چاپگر, شناسایی منبع چاپ, فضای متغیر کل چاپگر, تغییرات درون کلاسی, بردار هویت,
چکیده مقاله :
پیشرفت در دنیای دیجیتال، ما را به سمت توسعه ابزار بازجویی قانونی دیجیتال سوق می دهد. استفاده از روش های یادگیری ماشین برای شناسایی منبع چاپ یکی از زیر مجموعه های این حوزه بوده که درحال توسعه است. در این مقاله، روش جدیدی برای استخراج ویژگی های ثانویه بر پایه بردار هویت (i-vector) برای شناسایی منبع چاپ ارائه شده است. در روش پیشنهادی تنها با استخراج یک بردار i-vector با بعد کم به‌ازای هر صفحه بدون استفاده از روش بازشناسی نوری نویسه ها (OCR) و با حذف رأی‌گیری اکثریت فرایند طبقه‌بندی تسریع شده است. به‌این ترتیب روش پیشنهادی در استخراج ویژگی ها مستقل از نوع و اندازه قلم نویسه‌ها و زبان متن است. ویژگی های ثانویه با افراز تصویر سند به تکه های کوچک‌تر و مدل سازی ویژگی‌های اولیه الگوی دودویی محلی (LBP) مربوط به ناحیه های تیره، مرز و روشن در فضاهای مجزا به دست می آید. مدل سازی ویژگی های اولیه نواحی مختلف در فضاهای مجزای متغیر کل چاپگر، امکان استخراج اطلاعات جداکننده کلاس ها از بافت چاپ باقیمانده در ناحیه روشن را برای افزایش دقت و صحت طبقه‌بندی مهیا می کند. در این مقاله تأثیر استفاده از بافت نواحی مختلف و تغییر ابعاد تکه بندی با استفاده از طبقه‌بند ماشین بردار پشتیبان (SVM) از طریق شبیه‌سازی به‌دقت بررسی شده است. نتایج شبیه سازی، نشان می دهد که تنها با پالایش ویژگی های اولیه LBP به صحت 05/99 درصد دست یافته ایم که بیشتر از آخرین پژوهش های این حوزه است.
Advances in the digital world are leading us to the development of digital forensic tools. The use of machine learning methods for source printer identification is one of the sub-fields of this area that is being developed. In this paper, a new method for extracting secondary features based on identity vector or i-vector to identify the print source is presented. In the proposed method, the classification process is accelerated only by extracting a low-dimension i-vector vector per page, without the use of optical character recognition (OCR) method, and by eliminating majority voting. Furthermore, the proposed method in extracting features is independent of the type and size of the font and the language of the text. Secondary features are obtained by splitting the document image into smaller patches and modeling the primary LBP features of the dark, border, and light areas in separate spaces. Modeling the primary features of different regions in separate total variability printer space makes it possible to extract class discriminator information from the remaining print texture in the bright area to increase classification accuracy. In this paper, the effect of using the texture of different regions and changing the patch dimensions using the SVM (Support Vector Machine) classifier through simulation has been carefully investigated. The simulation results show that only by refining the basic features of LBP we achieved 99.05% accuracy, which is more than the latest research in this field.
[1] P. Yang, D. Baracchi, R. Ni, Y. Zhao, F. Argenti, A. Piva, “A survey of deep learning-based source image forensics”, Journal Imaging, vol. 6, no. 3, p. 9, Mar. 2020 (doi: 10.3390/jimaging6030009).
[2] V. Itier, O. Strauss, L. Morel, W. Puech, “Color noise correlation-based splicing detection for image forensics”, Multimedia Tools and Applications, pp. 1–19, Jan. 2021 (doi: 10.1007/s11042-020-10326-5).
[3] A.T.S. Ho, S. Li, Handbook of digital forensics of multimedia data and devices, Chichester, UK: John Wiley & Sons, Ltd, 2015.
[4] P.-J. Chiang, J.P. Allebach, G.T.-C. Chiu, “Extrinsic signature embedding and detection in electrophotographic halftoned images through exposure modulation”, IEEE Trans. on Information Forensics and Security, vol. 6, no. 3, pp. 946–959, Sept. 2011 (doi: 10.1109/TIFS.2011.2156789).
[5] P.J. Chiang, G.N. Ali, A.K. Mikkilineni, E.J. Delp, J.P. Allebach, G.T.C. Chiu, “Extrinsic signatures embedding and detection for information hiding and secure printing in electrophotography”, Proceeding of the IEEE/ACC, pp. 1-6, Minneapolis, MN, USA, June 2006 (doi: 10.1109/ACC.2006.1656604).
[6] G. Adams, S. Pollard, S. Simske, “A study of the interaction of paper substrates on printed forensic imaging”, Proceedings of the ACM, pp. 263-266, Limerick , Ireland,Sept. 2011 (doi: 10.1145/2034691.2034743).
[7] W. Jiang, A.T.S.S. Ho, H. Treharne, Y.Q. Shi, “A novel multi-size block Benford’s law scheme for printer identification”, Pacific-Rim Conference on Multimedia, vol. 6297 LNCS, no. PART 1, pp. 643–652, Springer, 2010.
[8] S. Joshi, N. Khanna, “Source printer classification using printer specific local texture descriptor”, IEEE Trans. on Information Forensics and Security, vol. 15, no. 1, pp. 160–171, 2020 (doi: 10.1109/TIFS.2019.2919869).
[9] S. Joshi, N. Khanna, “Single classifier-based passive system for source printer classification using local texture features”, IEEE Trans. on Information Forensics and Security, vol. 13, no. 7, pp. 1603–1614, July 2018 (doi: 10.1109/TIFS.2017.2779441).
[10] A. Ferreira, L. Bondi, L. Baroffio, P. Bestagini, J. Huang, J.A. Santos, S. Tubaro, A. Rocha, “Data-driven feature characterization techniques for laser printer attribution”, IEEE Trans. on Information Forensics and Security, vol. 12, no. 8, pp. 1860–1873, Aug. 2017 (doi: 10.1109/TIFS.2017.2692722).
[11] L.C. Navarro, A.K.W. Navarro, A. Rocha, R. Dahab, “Connecting the dots: Toward accountable machine-learning printer attribution methods”, Journal of Visual Communication and Image Representation, vol. 53, pp. 257–272, May 2018 (doi: 10.1016/j.jvcir.2018.04.002).
[12] A. Ferreira, L.C. Navarro, G. Pinheiro, J.A. Santos, A. Rocha, “Laser printer attribution: Exploring new features and beyond”, Forensic Science International, vol. 247, pp. 105–125, Feb. 2015 (doi: 10.1016/j.forsciint.2014.11.030).
[13] M.J. Tsai, I. Yuadi, Y.H. Tao, “Decision-theoretic model to identify printed sources”, Multimedia Tools and Applications, vol. 77, no. 20, pp. 27543–27587, Oct. 2018 (doi: 10.1007/s11042-018-5938-0).
[14] J. Hao, X. Kong, S. Shang, “Printer identification using page geometric distortion on text lines”, Proceeding of the IEEE/ChinaSIP, pp. 856–860, Chengdu, China, July 2015 (doi: 10.1109/ChinaSIP.2015.7230526).
[15] P.J. Chiang, N. Khanna, A.K. Mikkilineni, M.V.O. Segovia, J.P. Allebach, G.T.C. Chiu, E.J. Delp, "Printer and scanner forensics: Models and methods”, Intelligent Multimedia Analysis for Security Applications, vol. 282, pp. 145–187, March 2010 (doi: 10.1007/978-3-642-11756-5_7).
[16] S. Escher, T. Strafe, “Robustness analysis of a passive printer identification scheme for halftone images”, Proceeding of the IEEE/ICIP, pp. 4357–4361, Beijing, China, Sept. 2017 (doi: 10.1109/ICIP.2017.8297105).
[17] P. Kenny, G. Boulianne, P. Ouellet, P. Dumouchel, “Joint factor analysis versus eigenchannels in speaker recognition”, IEEE Trans. on Audio, Speech and Language Processing, vol. 15, no. 4, pp. 1435–1447, May 2007 (doi: 10.1109/TASL.2006.881693).
[18] P. Kenny, P. Ouellet, N. Dehak, V. Gupta, P. Dumouchel, “A study of interspeaker variability in speaker verification”, IEEE Trans. on Audio, Speech and Language Processing, vol. 16, no. 5, pp. 980–988, July 2008 (doi: 10.1109/TASL.2008.925147).
[19] N. Dehak, P. J. Kenny, R. Dehak, P. Dumouchel, P. Ouellet, “Front-end factor analysis for speaker verification”, IEEE Trans. on Audio, Speech and Language Processing, vol. 19, no. 4, pp. 788–798, May 2011 (doi: 10.1109/TASL.2010.2064307).
[20] P. Verma, P.K. Das, “I-Vectors in speech processing applications: a survey”, International Journal of Speech Technology, vol. 18, no. 4, pp. 529–546, Dec. 2015 (doi: 10.1007/s10772-015-9295-3).
[21] Y. Xing, P. Tan, C. Zhang, “Improved i-vector speaker verification based on WCCN and ZT-norm”, Chinese Conference on Biometric Recognition, pp. 424-431, Springer, Cham, Oct. 2016 (doi: 10.1007/978-3-319-46654-5_47).
[22] T. Ojala, M. Pietikäinen, D. Harwood, “A comparative study of texture measures with classification based on featured distributions”, Pattern Recognition, vol. 29, no. 1, pp. 51–59, Jan. 1996 (doi: 10.1016/0031-3203(95)00067-4).
_||_[1] P. Yang, D. Baracchi, R. Ni, Y. Zhao, F. Argenti, A. Piva, “A survey of deep learning-based source image forensics”, Journal Imaging, vol. 6, no. 3, p. 9, Mar. 2020 (doi: 10.3390/jimaging6030009).
[2] V. Itier, O. Strauss, L. Morel, W. Puech, “Color noise correlation-based splicing detection for image forensics”, Multimedia Tools and Applications, pp. 1–19, Jan. 2021 (doi: 10.1007/s11042-020-10326-5).
[3] A.T.S. Ho, S. Li, Handbook of digital forensics of multimedia data and devices, Chichester, UK: John Wiley & Sons, Ltd, 2015.
[4] P.-J. Chiang, J.P. Allebach, G.T.-C. Chiu, “Extrinsic signature embedding and detection in electrophotographic halftoned images through exposure modulation”, IEEE Trans. on Information Forensics and Security, vol. 6, no. 3, pp. 946–959, Sept. 2011 (doi: 10.1109/TIFS.2011.2156789).
[5] P.J. Chiang, G.N. Ali, A.K. Mikkilineni, E.J. Delp, J.P. Allebach, G.T.C. Chiu, “Extrinsic signatures embedding and detection for information hiding and secure printing in electrophotography”, Proceeding of the IEEE/ACC, pp. 1-6, Minneapolis, MN, USA, June 2006 (doi: 10.1109/ACC.2006.1656604).
[6] G. Adams, S. Pollard, S. Simske, “A study of the interaction of paper substrates on printed forensic imaging”, Proceedings of the ACM, pp. 263-266, Limerick , Ireland,Sept. 2011 (doi: 10.1145/2034691.2034743).
[7] W. Jiang, A.T.S.S. Ho, H. Treharne, Y.Q. Shi, “A novel multi-size block Benford’s law scheme for printer identification”, Pacific-Rim Conference on Multimedia, vol. 6297 LNCS, no. PART 1, pp. 643–652, Springer, 2010.
[8] S. Joshi, N. Khanna, “Source printer classification using printer specific local texture descriptor”, IEEE Trans. on Information Forensics and Security, vol. 15, no. 1, pp. 160–171, 2020 (doi: 10.1109/TIFS.2019.2919869).
[9] S. Joshi, N. Khanna, “Single classifier-based passive system for source printer classification using local texture features”, IEEE Trans. on Information Forensics and Security, vol. 13, no. 7, pp. 1603–1614, July 2018 (doi: 10.1109/TIFS.2017.2779441).
[10] A. Ferreira, L. Bondi, L. Baroffio, P. Bestagini, J. Huang, J.A. Santos, S. Tubaro, A. Rocha, “Data-driven feature characterization techniques for laser printer attribution”, IEEE Trans. on Information Forensics and Security, vol. 12, no. 8, pp. 1860–1873, Aug. 2017 (doi: 10.1109/TIFS.2017.2692722).
[11] L.C. Navarro, A.K.W. Navarro, A. Rocha, R. Dahab, “Connecting the dots: Toward accountable machine-learning printer attribution methods”, Journal of Visual Communication and Image Representation, vol. 53, pp. 257–272, May 2018 (doi: 10.1016/j.jvcir.2018.04.002).
[12] A. Ferreira, L.C. Navarro, G. Pinheiro, J.A. Santos, A. Rocha, “Laser printer attribution: Exploring new features and beyond”, Forensic Science International, vol. 247, pp. 105–125, Feb. 2015 (doi: 10.1016/j.forsciint.2014.11.030).
[13] M.J. Tsai, I. Yuadi, Y.H. Tao, “Decision-theoretic model to identify printed sources”, Multimedia Tools and Applications, vol. 77, no. 20, pp. 27543–27587, Oct. 2018 (doi: 10.1007/s11042-018-5938-0).
[14] J. Hao, X. Kong, S. Shang, “Printer identification using page geometric distortion on text lines”, Proceeding of the IEEE/ChinaSIP, pp. 856–860, Chengdu, China, July 2015 (doi: 10.1109/ChinaSIP.2015.7230526).
[15] P.J. Chiang, N. Khanna, A.K. Mikkilineni, M.V.O. Segovia, J.P. Allebach, G.T.C. Chiu, E.J. Delp, "Printer and scanner forensics: Models and methods”, Intelligent Multimedia Analysis for Security Applications, vol. 282, pp. 145–187, March 2010 (doi: 10.1007/978-3-642-11756-5_7).
[16] S. Escher, T. Strafe, “Robustness analysis of a passive printer identification scheme for halftone images”, Proceeding of the IEEE/ICIP, pp. 4357–4361, Beijing, China, Sept. 2017 (doi: 10.1109/ICIP.2017.8297105).
[17] P. Kenny, G. Boulianne, P. Ouellet, P. Dumouchel, “Joint factor analysis versus eigenchannels in speaker recognition”, IEEE Trans. on Audio, Speech and Language Processing, vol. 15, no. 4, pp. 1435–1447, May 2007 (doi: 10.1109/TASL.2006.881693).
[18] P. Kenny, P. Ouellet, N. Dehak, V. Gupta, P. Dumouchel, “A study of interspeaker variability in speaker verification”, IEEE Trans. on Audio, Speech and Language Processing, vol. 16, no. 5, pp. 980–988, July 2008 (doi: 10.1109/TASL.2008.925147).
[19] N. Dehak, P. J. Kenny, R. Dehak, P. Dumouchel, P. Ouellet, “Front-end factor analysis for speaker verification”, IEEE Trans. on Audio, Speech and Language Processing, vol. 19, no. 4, pp. 788–798, May 2011 (doi: 10.1109/TASL.2010.2064307).
[20] P. Verma, P.K. Das, “I-Vectors in speech processing applications: a survey”, International Journal of Speech Technology, vol. 18, no. 4, pp. 529–546, Dec. 2015 (doi: 10.1007/s10772-015-9295-3).
[21] Y. Xing, P. Tan, C. Zhang, “Improved i-vector speaker verification based on WCCN and ZT-norm”, Chinese Conference on Biometric Recognition, pp. 424-431, Springer, Cham, Oct. 2016 (doi: 10.1007/978-3-319-46654-5_47).
[22] T. Ojala, M. Pietikäinen, D. Harwood, “A comparative study of texture measures with classification based on featured distributions”, Pattern Recognition, vol. 29, no. 1, pp. 51–59, Jan. 1996 (doi: 10.1016/0031-3203(95)00067-4).