Comparing Feature Matching Methods to Identify Persian Writers

Document Type : Original Article


1 Department of Computer Science, Ferdows Branch, Islamic Azad University Ferdows, Iran

2 Assistant Professor, Department of Computer Engineering, Faculty of Engineering, Islamic Azad University Ferdows Branch, Iran


A variety of frequently utilized feature matching techniques are employed and integrated in this study to tackle the task of writer identification using Persian scripts. The literature demonstrates promise for feature matching methods in the writer identification field. Multiple methods, such as SIFT, SURF, BRISK, BRISK-SURF, FREAK, SURF-FREAK, SURF-BRISK, Harris-SURF, Harris-FREAK, and Harris-BRISK, were employed and their performance was analyzed and compared to establish the optimal method(s). The identification process involves the comparison of the query image with all reference images, and as such, the reference image with the highest number of matched feature points corresponds to the query image. To the best of our knowledge, no extensive study has compared the effectiveness of these methods in the context of identifying Persian writers. Among the methods compared, SIFT and SURF algorithms have demonstrated the highest accuracy in recognizing the correct writer, while other algorithms have exhibited encouraging results.


  • Aliakbarzadeh, M., & Razzazi, F. (2020). Online Persian/Arabic Writer Identification using Gated Recurrent Unit Neural Networks. Majlesi Journal of Electrical Engineering, 14, 73-79.
  • Khosravi, S., & Chalechale, A. (2022). Recognition of Persian/Arabic Handwritten Words Using a Combination of Convolutional Neural Networks and Autoencoder (AECNN). Mathematical Problems in Engineering.
  • Safarzadeh, V., & Jafarzadeh, P. (2020). Offline Persian Handwriting Recognition with CNN and RNN-CTC. 2020 25th International Computer Conference, Computer Society of Iran (CSICC), 1-10.
  • Kumar, R., Chanda, B., & Sharma, J. D. (2014). A novel sparse model based forensic writer identification. Pattern Recognition Letters, 35, 105–112. doi:10.1016/j.patrec.2013.07.001.
  • Schomaker, L., Franke, K., & Bulacu, M. (2007). Using codebooks of fragmented connected-component contours in forensic and historic writer identification. Pattern Recognition Letters, 28(6), 719–727. doi:10.1016/j.patrec.2006.08.005.
  • Said, H. E., Tan, T. N., & Baker, K. D. (2000). Personal identification based on handwriting. Pattern Recognition, 33(1), 149-160.
  • Hannad, Y., Siddiqi, I., & El Kettani, M. E. Y. (2016). Writer identification using texture descriptors of handwritten fragments. Expert Systems with Applications, 47, 14–22. doi:10.1016/j.eswa.2015.11.002.
  • Peyvandi, M., & Zahedi, M. (2013). 11th Iranian Conference on Intelligent Systems. (In Persian).
  • Wu, X., Tang, Y., & Bu, W. (2014). Offline text-independent writer identification based on scale invariant feature transform. IEEE Transactions on Information Forensics and Security, 9(3), 526–536. doi:10.1109/tifs.2014.2301274.
  • Su, T., Zhang, T., & Guan, D. (2007). Corpus-based HIT-MW database for offline recognition of general-purpose Chinese handwritten text. International Journal on Document Analysis and Recognition, 10(1), 27–38. doi:10.1007/s10032-006-0037-6.
  • Bulacu, M., Schomaker, L., & Vuurpijl, L. (2005). Writer identification using edge-based directional features. Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings. Presented at the Seventh International Conference on Document Analysis and Recognition, Edinburgh, UK. doi:10.1109/icdar.2003.1227797.
  • Hassaine, A., & Maadeed, S. A. (2012). ICFHR 2012 competition on writer identification challenge 2: Arabic scripts. 2012 International Conference on Frontiers in Handwriting Recognition. Presented at the 2012 International Conference on Frontiers in Handwriting Recognition (ICFHR), Bari, Italy. doi:10.1109/icfhr.2012.218.
  • Marti, U.-V., & Bunke, H. (2002). The IAM-database: an English sentence database for offline handwriting recognition. International Journal on Document Analysis and Recognition, 5(1), 39–46. doi:10.1007/s100320200071.
  • Sharma, M. K., & Dhaka, V. P. (2015). Offline language-free writer identification based on speeded-up robust features. International Journal of Engineering-Transactions A: Basics, 28(7), 984–994.
  • Harris, C., & Stephens, M. (1988). A combined corner and edge detector. Procedings of the Alvey Vision Conference 1988. Presented at the Alvey Vision Conference 1988, Manchester. doi:10.5244/c.2.23.
  • Ben-Musa, A. S., Singh, S. K., & Agrawal, P. (2014, July). Object detection and recognition in cluttered scene using Harris Corner Detection. 2014 International Conference on Control, Instrumentation, Communication and Computational Technologies (ICCICCT). India. doi:10.1109/iccicct.2014.6992953.
  • Bhowmik, M. K., Shil, S., & Saha, P. (2013). Feature points extraction of thermal face using Harris interest point detection. Procedia Technology, 10, 724–730. doi:10.1016/j.protcy.2013.12.415.
  • Rosten, E., & Drummond, T. (2006). Machine learning for high-speed corner detection. In Lecture Notes in Computer Science. Computer Vision – ECCV 2006 (pp. 430–443). doi:10.1007/11744023_34.
  • Alahi, A., Ortiz, R., & Vandergheynst, P. (2012). FREAK: Fast Retina Keypoint. 2012 IEEE Conference on Computer Vision and Pattern Recognition. Presented at the 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Providence, RI. doi:10.1109/cvpr.2012.6247715.
  • Leutenegger, S., Chli, M., & Siegwart, R. Y. (2011). BRISK: Binary Robust invariant scalable keypoints. 2011 International Conference on Computer Vision. Presented at the 2011 IEEE International Conference on Computer Vision (ICCV), Barcelona, Spain. doi:10.1109/iccv.2011.6126542.
  • Keyvanpour, M., Tavoli, R., & Mozafari, S. (2014). Document image retrieval based on keyword spotting using relevance feedback.
  • Rusinol, M., Aldavert, D., Toledo, R., & Llados, J. (2011). Browsing heterogeneous document collections by a segmentation-free word spotting method. 2011 International Conference on Document Analysis and Recognition. Beijing, China. doi:10.1109/icdar.2011.22.
  • Madbouly, A. M. M., Wafy, M., & Mostafa, M. S. M. (2015). Performance Assessment of Feature Detector-Descriptor Combination. International Journal of Computer Science Issues (IJCSI), 12(5).
  • Otsu, N. (1979). A threshold selection method from gray-level histograms. IEEE transactions on systems, man, and cybernetics, 9(1), 62-66.
  • Lam, L., Lee, S.-W., & Suen, C. Y. (1992). Thinning methodologies-a comprehensive survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 14(9), 869–885. doi:10.1109/34.161346
  • Lowe, D. G. (2004). Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision, 60(2), 91–110. doi:10.1023/b:visi.0000029664.99615.94.
  • Bay, H., Ess, A., Tuytelaars, T., & Van Gool, L. (2008). Speeded-up robust features (SURF). Computer vision and image understanding, 110(3), 346-359.
  • Javadzadeh, R. (2017). Persian Writer Identification Dataset.
  • Vedaldi, A., & Fulkerson, B. (2010, October). VLFeat: An open and portable library of computer vision algorithms. In Proceedings of the 18th ACM international conference on Multimedia (pp. 1469-1472). ACM.