Skip to main content

Feature learning and encoding for multi-script writer identification

International Journal on Document Analysis and Recognition (IJDAR) volume 25, pages 79–93 (2022)Cite this article

Abstract

Writer identification from handwriting samples has been an interesting research problem for the pattern recognition community in general and handwriting recognition community in particular. In most cases, however, it is assumed that writers produce writing samples in a single script only. A more challenging scenario is the multi-script writer identification where the training and test samples of writers belong to different scripts. This paper presents a deep learning-based solution for writer identification in a multi-script scenario. The technique relies on identifying keypoints in handwriting and extracting small patches around these keypoints. These patches are aimed to capture the writing gestures of individuals which are likely to be common across multiple scripts. Robust feature representations are learned from these patches using a deep convolutional neural network and the features are encoded using a newly proposed variant of the Vector of Locally Aggregated Descriptors (VLAD). Experiments on three bilingual handwriting datasets including writing samples in Arabic, English, French, Chinese and Farsi report promising identification rates and significantly outperform the current state-of-the-art on this problem.

This is a preview of subscription content, access via your institution.

Access options

Buy single article

Instant access to the full article PDF.

USD 39.95

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

References

  1. Abbas, Faycel, Gattal, Abdeljalil, Djeddi, Chawki, Siddiqi, Imran, Bensefia, Ameur, Saoudi, Kamel: Texture feature column scheme for single-and multi-script writer identification. IET Biometr. 10(2), 179–193 (2021)

    Article  Google Scholar 

  2. Gattal Abdeljalil, Chawki Djeddi, Imran Siddiqi, and Somaya Al-Maadeed. Writer identification on historical documents using oriented basic image features. In 2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR), pages 369–373. IEEE, 2018

  3. Mohamed Nidhal Abdi and Maher Khemakhem: A model-based approach to offline text-independent arabic writer identification and verification. Pattern Recognit. 48(5), 1890–1903 (2015)

    Article  Google Scholar 

  4. Félix Abecassis. Opencv-morphological skeleton. Retrieved from Félix Abecassis Projects and Experiments: International Journal of Remote Sensinghttp://felix.abecassis.me/2011/09/opencv-morphological-skeleton/geological mapping at Cuprite Nevada:a rule-based system, 31:7, 2011

  5. Somaya Al-Maadeed, Abdelaali Hassaine, Ahmed Bouridane, and Muhammad Atif Tahir. Novel geometric features for off-line writer identification. Pattern Analysis and Applications, 19(3):699–708, 2016

  6. Bennour, Akram, Djeddi, Chawki, Gattal, Abdeljalil, Siddiqi, Imran, Mekhaznia, Tahar: Handwriting based writer recognition using implicit shape codebook. Forensic Sci. Int. 301, 91–100 (2019)

    Article  Google Scholar 

  7. Ameur Bensefia, Ali Nosary, Thierry Paquet, and Laurent Heutte. Writer identification by writer’s invariants. In: Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition, pages 274–279. IEEE, 2002

  8. Bensefia, Ameur, Paquet, Thierry, Heutte, Laurent: A writer identification and verification system. Pattern Recogonit Lett. 26(13), 2080–2092 (2005)

    Article  Google Scholar 

  9. Bertolini, Diego, Oliveira, Luiz S., Justino, E., Sabourin, Robert: Texture-based descriptors for writer identification and verification. Expert Syst. with Appl. 40(6), 2069–2080 (2013)

    Article  Google Scholar 

  10. Bulacu, Marius, Schomaker, Lambert: Text-independent writer identification and verification using textural and allographic features. Pattern Anal. Mach. Intell. IEEE Trans 29(4), 701–717 (2007)

    Article  Google Scholar 

  11. Djeddi Chawki and Souici-Meslati Labiba. A texture based approach for arabic writer identification and verification. In: 2010 International Conference on Machine and Web Intelligence, pages 115–120. IEEE, 2010

  12. Vincent Christlein, David Bernecker, and Elli Angelopoulou. Writer identification using vlad encoded contour-zernike moments. In: 2015 13th International Conference on Document Analysis and Recognition (ICDAR), pages 906–910. IEEE, 2015

  13. Vincent Christlein, David Bernecker, Andreas Maier, and Elli Angelopoulou. Offline writer identification using convolutional neural network activation features. In: German Conference on Pattern Recognition, pages 540–552. Springer, 2015

  14. Vincent Christlein, Martin Gropp, Stefan Fiel, and Andreas Maier. Unsupervised feature learning for writer identification and writer retrieval. In: 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), volume 1, pages 991–997. IEEE, 2017

  15. Vincent Christlein and Andreas Maier. Encoding cnn activations for writer recognition. In:D 2018 13th IAPR International Workshop on Document Analysis Systems (DAS), pages 169–174. IEEE, 2018

  16. Jonathan Delhumeau, Philippe-Henri Gosselin, Hervé Jégou, and Patrick Pérez. Revisiting the vlad image representation. In: Proceedings of the 21st ACM international conference on Multimedia, pages 653–656, 2013

  17. Chawki Djeddi, Somaya Al-Maadeed, Abdeljalil Gattal, Imran Siddiqi, Abdellatif Ennaji, and Haikal El Abed. Icfhr2016 competition on multi-script writer demographics classification using” quwi” database. In: 2016 15th International Conference on Frontiers in Handwriting Recognition (ICFHR), pages 602–606. IEEE, 2016

  18. Chawki Djeddi, Somaya Al-Maadeed, Abdeljalil Gattal, Imran Siddiqi, Labiba Souici-Meslati, and Haikal El Abed. Icdar2015 competition on multi-script writer identification and gender classification using ‘quwi’database. In: 2015 13th International Conference on Document Analysis and Recognition (ICDAR), pages 1191–1195. IEEE, 2015

  19. Chawki Djeddi, Somaya Al-Maadeed, Imran Siddiqi, Gattal Abdeljalil, Sheng He, and Younes Akbari. Icfhr 2018 competition on multi-script writer identification. In: 2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR), pages 506–510. IEEE, 2018

  20. Chawki Djeddi, Abdeljalil Gattal, Labiba Souici-Meslati, Imran Siddiqi, Youcef Chibani, and Haikal El Abed. Lamis-mshd: a multi-script offline handwriting database. In: 2014 14th International Conference on Frontiers in Handwriting Recognition, pages 93–97. IEEE, 2014

  21. Chawki Djeddi, Imran Siddiqi, Labiba Souici-Meslati, and Abdellatif Ennaji. Multi-script writer identification optimized with retrieval mechanism. In: 2012 International Conference on Frontiers in Handwriting Recognition, pages 509–514. IEEE, 2012

  22. Djeddi, Chawki, Siddiqi, Imran, Souici-Meslati, Labiba, Ennaji, Abdellatif: Text-independent writer recognition using multi-script handwritten texts. Pattern Recognit. Lett. 34(10), 1196–1202 (2013)

    Article  Google Scholar 

  23. bibitemfecker2014writer D Fecker, A Asit, Volker Märgner, Jihad El-Sana, and Tim Fingscheidt. Writer identification for historical arabic documents. In: 2014 22nd International Conference on Pattern Recognition, pages 3050–3055. IEEE, 2014

  24. Stefan Fiel and Robert Sablatnig. Writer identification and retrieval using a convolutional neural network. In: International Conference on Computer Analysis of Images and Patterns, pages 26–37. Springer, 2015

  25. Utpal Garain and Thierry Paquet. Off-line multi-script writer identification using ar coefficients. In: 2009 10th International Conference on Document Analysis and Recognition, pages 991–995. IEEE, 2009

  26. Ghiasi, Golnaz, Safabakhsh, Reza: Offline text-independent writer identification using codebook and efficient code extraction methods. Image Vision Comput. 31(5), 379–391 (2013)

    Article  Google Scholar 

  27. Tara Gilliam, Richard C Wilson, and John A Clark. Scribe identification in medieval english manuscripts. In: 2010 20th International Conference on Pattern Recognition, pages 1880–1883. IEEE, 2010

  28. Guo, Zhenhua, Zhang, Lei, Zhang, David: A completed modeling of local binary pattern operator for texture classification. IEEE Trans. Image Process. 19(6), 1657–1663 (2010)

    Article  MathSciNet  Google Scholar 

  29. Yaâcoub Hannad, Imran Siddiqi, Chawki Djeddi, and Mohamed El-Youssfi El-Kettani. Improving arabic writer identification using score-level fusion of textural descriptors. IET Biometrics, 8(3):221–229, 2019

  30. Hannad, Yaacoub, Siddiqi, Imran, El Youssfi, Mohamed, Kettani, El.: Writer identification using texture descriptors of handwritten fragments. Expert Syst. Appl. 47, 14–22 (2016)

    Article  Google Scholar 

  31. Christopher G Harris, Mike Stephens, et al. A combined corner and edge detector. In: Alvey vision conference, volume 15, pages 10–5244. Citeseer, 1988

  32. Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016

  33. Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Identity mappings in deep residual networks. In: European conference on computer vision, pages 630–645. Springer, 2016

  34. He, Sheng, Wiering, Marco, Schomaker, Lambert: Junction detection in handwritten documents and its application to writer identification. Pattern Recognit. 48(12), 4036–4048 (2015)

    Article  Google Scholar 

  35. Zhenyu He, Xinge You, and Yuan Yan Tang. Writer identification using global wavelet-based features. Neurocomputing, 71(10-2):1832–1841, 2008

  36. Rajiv Jain and David Doermann. Offline writer identification using k-adjacent segments. In: 2011 International Conference on Document Analysis and Recognition, pages 769–773. IEEE, 2011

  37. Hervé Jégou, Matthijs Douze, and Cordelia Schmid. On the burstiness of visual elements. In: 2009 IEEE conference on computer vision and pattern recognition, pages 1169–1176. IEEE, 2009

  38. Jegou, Herve, Perronnin, Florent, Douze, Matthijs, Sánchez, Jorge, Perez, Patrick, Schmid, Cordelia: Aggregating local image descriptors into compact codes. IEEE Trans. Pattern Anal. Mach. Intell. 34(9), 1704–1716 (2011)

    Article  Google Scholar 

  39. Tak-Eun Kim and Myoung Ho Kim: Improving the search accuracy of the vlad through weighted aggregation of local descriptors. J. Visual Comm. Image Represent. 31, 237–252 (2015)

    Article  Google Scholar 

  40. Neeraj Kumar, Li Zhang, and Shree Nayar. What is a good nearest neighbors algorithm for finding similar patches in images? In:D European conference on computer vision, pages 364–378. Springer, 2008

  41. Lai, Songxuan, Zhu, Yecheng, Jin, Lianwen: Encoding pathlet and sift features with bagged vlad for historical writer identification. IEEE Trans. Inf. Forensics Secur. 15, 3553–3566 (2020)

    Article  Google Scholar 

  42. Georgios Louloudis, Basilis Gatos, and Nikolaos Stamatopoulos. Icfhr 2012 competition on writer identification challenge 1: Latin/greek documents. In: 2012 International Conference on Frontiers in Handwriting Recognition, pages 829–834. IEEE, 2012

  43. Alieh Masomi, Hamid Reza Ghafari, Kazem Nouri, Younes Akbari, Walid Bouamra, and Chawki Djeddi. A new database for writer demographics attributes detection based on off-line persian and english handwriting. In: Proceedings of the Mediterranean Conference on Pattern Recognition and Artificial Intelligence, pages 125–130, 2016

  44. Andrew J Newell and Lewis D Griffin. Writer identification using oriented basic image features and the delta encoding. Pattern Recognit., 47(6):2255–2265, 2014

  45. Nguyen, Hung Tuan, Nguyen, Cuong Tuan, Ino, Takeya, Indurkhya, Bipin, Nakagawa, Masaki: Text-independent writer identification using convolutional neural network. Pattern Recognit. Lett. 121, 104–112 (2019)

    Article  Google Scholar 

  46. Stephen M Omohundro. Five balltree construction algorithms. International Computer Science Institute Berkeley, 1989

  47. Florent Perronnin, Jorge Sánchez, and Thomas Mensink. Improving the fisher kernel for large-scale image classification. In: European conference on computer vision, pages 143–156. Springer, 2010

  48. Arshia Rehman, Saeeda Naz, Muhammad Imran Razzak, and Ibrahim A Hameed. Automatic visual features for writer identification: A deep learning approach. IEEE access, 7:17149–17157, 2019

  49. Huwida ES Said, Tienniu N Tan, and Keith D Baker. Personal identification based on handwriting. Pattern Recognition, 33(1):149–160, 2000

  50. Lambert Schomaker. Advances in writer identification and verification. In: Ninth International Conference on Document Analysis and Recognition (ICDAR 2007), volume 2, pages 1268–1273. IEEE, 2007

  51. Schomaker, Lambert, Bulacu, Marius: Automatic writer identification using connected-component contours and edge-based features of uppercase western script. IEEE Transactions on Pattern Analysis and Machine Intelligence 26(6), 787–798 (2004)

    Article  Google Scholar 

  52. Abdelillah Semma, Yaâcoub Hannad, and Mohamed El Youssfi El Kettani. Impact of the cnn patch size in the writer identification. In: Networking, Intelligent Systems and Security, pages 103–114. Springer, 2022

  53. Semma, Abdelillah, Hannad, Yaâcoub., Siddiqi, Imran, Djeddi, Chawki, El Youssfi, Mohamed, Kettani, El (2021)Writer identification using deep learning with fast keypoints and harris corner detector. Expert Syst. Appl. 184, 115473

  54. Semma, Abdelillah, Lazrak, Said, Hannad, Yaâcoub., Boukhani, Mohamed, El Kettani, Youssfi: Writer identification: The effect of image resizing on cnn performance. The Int. Archives . Photogramm. Remote Sens. Spatial Inf. Sci 46, 501–507 (2021)

    Article  Google Scholar 

  55. Sheng, Biyun, Shen, Chunhua, Lin, Guosheng, Li, Jun, Yang, Wankou, Sun, Changyin: Crowd counting via weighted vlad on a dense attribute feature map. IEEE Trans. Circuits Syst. Video Techno. 28(8), 1788–1797 (2016)

    Article  Google Scholar 

  56. Imran Siddiqi and Nicole Vincent. Writer identification in handwritten documents. In: Ninth International Conference on Document Analysis and Recognition (ICDAR 2007), volume 1, pages 108–112. IEEE, 2007

  57. Siddiqi, Imran, Vincent, Nicole: Text independent writer recognition using redundant writing patterns with contour-based orientation and curvature features. Pattern Recognit. 43(11), 3853–3865 (2010)

    Article  Google Scholar 

  58. Sargur N Srihari, Sung-Hyuk Cha, Hina Arora, and Sangjik Lee. Individuality of handwriting. J. Forensic Sci., 47(4):856–872, 2002

  59. Guo Xian Tan, Christian Viard-Gaudin, and Alex C Kot. Individuality of alphabet knowledge in online writer identification. In: International Journal on Document Analysis and Recognition (IJDAR), 13(2):147–157, 2010

  60. Yanhong Wang, Yigang Cen, Liequan Liang, Linna Zhang, Viacheslav Voronin, and Vladimir Mladenovic. Fusion of deep features and weighted vlad vectors based on multiple features for image retrieval. In MATEC Web of Conferences, 2017

  61. Xiangqian, Wu., Tang, Youbao, Wei, Bu.: Offline text-independent writer identification based on scale invariant feature transform. IEEE Transactions on Information Forensics and Security 9(3), 526–536 (2014)

    Article  Google Scholar 

  62. Linjie Xing and Yu Qiao. Deepwriter: A multi-stream deep cnn for text-independent writer identification. I:n 2016 15th International Conference on Frontiers in Handwriting Recognition (ICFHR), pages 584–589. IEEE, 2016

  63. Yu-Jie Xiong, Ying Wen, Patrick SP Wang, and Yue Lu. Text-independent writer identification using sift descriptor and contour-directional feature. In 2015 13th International Conference on Document Analysis and Recognition (ICDAR), pages 91–95. IEEE, 2015

  64. Yang, Weixin, Jin, Lianwen, Liu, Manfei: Deepwriterid: an end-to-end online text-independent writer identification system. IEEE Intell. Syst. 31(2), 45–53 (2016)

    Article  MathSciNet  Google Scholar 

  65. Zhang, Xu-Yao., Xie, Guo-Sen., Liu, Cheng-Lin., Bengio, Yoshua: End-to-end online writer identification with recurrent neural network. IEEE Trans. Human–Mach. Syst. 47(2), 285–292 (2016)

    Article  Google Scholar 

  66. Yong Zhu, Tieniu Tan, and Yunhong Wang. Biometric personal identification based on handwriting. In: Proceedings 15th International Conference on Pattern Recognition. ICPR-2000, volume 2, pages 797–800. IEEE, 2000

Download references

Author information

Authors and Affiliations

  1. Ibn Tofail University, Kenitra, Morocco

    Abdelillah Semma, Said Lazrak & Mohamed El Youssfi El Kettani

  2. Faculty of Educational Sciences, Mohammed V University, Rabat, Morocco

    Yaâcoub Hannad

  3. Bahria University, Islamabad, Pakistan

    Imran Siddiqi

Authors
  1. Abdelillah Semma
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Yaâcoub Hannad
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Imran Siddiqi
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Said Lazrak
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Mohamed El Youssfi El Kettani
    View author publications

    You can also search for this author in PubMed Google Scholar

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Semma, A., Hannad, Y., Siddiqi, I. et al. Feature learning and encoding for multi-script writer identification. IJDAR 25, 79–93 (2022). https://doi.org/10.1007/s10032-022-00394-8

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10032-022-00394-8

Keywords

  • Multi-script writer Identification
  • Handwriting keypoints
  • Feature learning
  • Feature encoding