Abstract

Forensic Document Examination has become more diverse, which require authenticity or validation in determination of genuineness or non genuineness, to revealforgery, alteration, addition, deletion, and personal identification. Signature identification is a most challenging mission in the field of forensic questioned document examination. Its aim is to determine forged signatures by matching the unknown signature with known signature. Signature is a handwritten deception of a person that engraves on document as a proof of uniqueness. A person’s signature serves as a trademark. It is generally a person’s most common writing act and such is largely habitual. Signature of a person may be constituted of only letters, or with letters as well as non-letter patterns, or may be constituted only with non-letter patterns. In this modern computational era, computational approach to handwritten signature is more relevant due to its accuracy, less time consumption. This paper efforts to give computational software used for identification of signature samples. The objective of the study was to identify the natural variation occur to signature with the help of image analyzing software “Digimizer”. The efficiency of the proposed method is based on results of 100 writers with 4 signatures of each writer.

• 1.   Saran V, Kumar S. Computational method for forensic verification of offline signatures. IOSR J Comput Eng. 2013;14(1):81-3.
• 2.   Srihari SN. Computational methods for handwritten questioned document examination. Washington (DC): U.S. Department of Justice; 2010. Report No.: NCJ 234471.
• 3.   Dhieb T, Njah S, Benoureta H, Bouhlel MS, Ouarda W, Alimi AM. Towards a novel biometric system for forensic document examination. Pattern Recognit Lett. 2020;131:61-7.
• 4.   Parziale A, Rizzo AP, Fontanella F, Marcelli A. An interactive tool for forensic handwriting examination. In: Proceedings of the 14th International Conference on Frontiers in Handwriting Recognition (ICFHR); 2014 Sep 1-4; Crete, Greece. IEEE; 2014. p. 448-53.
• 5.   Bensefia A, Paquet T, Heutte L. Handwritten document analysis for automatic writer recognition. Electron Lett Comput Vis Image Anal. 2005;5(2):72-86.
• 6.   Srihari SN, Srinivasan H. Machine learning for signature verification. In: Franke K, Srihari SN, editors. Computational forensics. IWCF 2008. Lecture Notes in Computer Science, vol 5158. Berlin: Springer; 2008. p. 115-26.
• 7.   U.S. Department of Justice. Supporting documentation for proposed uniform language for testimony and reports for the forensic handwriting analysis discipline [Internet]. Washington (DC): Department of Justice; [cited 2026 Jan 21]. Srihari S, Leedham G. A survey of computer methods in forensic handwritten document examination. In: Proceedings of the 11th Conference of the International Graphonomics Society (IGS); 2003 Nov 2-5; Scottsdale, AZ. International Graphonomics Society; 2003. p. 278-81.
• 8.   Sanchez-Reillo R, Liu-Jimenez J. Forensic validation of biometrics using dynamic handwritten signatures. In: 2018 International Carnahan Conference on Security Technology (ICCST); 2018 Oct 22-25; Montreal, Canada. IEEE; 2018. p. 1-5.
• 9.   Mohammed LA. Forensic examination of signatures. London: Academic Press; 2019.
• 10.   Allen M. Foundations of forensic document analysis: theory and practice. Chichester (UK): Wiley-Blackwell; 2015.
• 11.   Hayes R. Forensic handwriting examination: a definitive guide [Internet]. Honolulu: ReedWrite Press; 2006 [cited 2026 Jan 21]. Chapter 1, Introduction to handwriting examination; [about 10 p.]. Wikipedia contributors. Signature [Internet]. Wikipedia, The Free Encyclopedia; [cited 2026 Jan 21].
• 12.   Srivastava A, Sharma K. Evaluation of common authorship characteristics in signatures and handwritings of different writers. J Emerg Technol Innov Res. 2020;7(5):26-32.