Handwritten Signature Verification Technology Using Deep Learning - A Review

Biometrics is currently widely utilized for the identification and verification of persons and their signatures all around the world. An individual's handwritten signature is a unique identifying work of human that is used and acknowledged primarily in banking and other financial and legal oper...

Full description

Saved in:
Bibliographic Details
Published in2022 3rd International Conference on Electronics and Sustainable Communication Systems (ICESC) pp. 813 - 817
Main Authors Tamrakar, Prasann, Badholia, Abhishek
Format Conference Proceeding
LanguageEnglish
Published IEEE 17.08.2022
Subjects
Banking
Biometrics (access control)
Communication systems
Deep learning
Feature Extraction
Forgery
Fraudulence
Handwriting recognition
Heuristic algorithms
Law
Signatures
Training
Verification
Online AccessGet full text

Cover

More Information
Summary:Biometrics is currently widely utilized for the identification and verification of persons and their signatures all around the world. An individual's handwritten signature is a unique identifying work of human that is used and acknowledged primarily in banking and other financial and legal operations. However, handwritten signatures are becoming increasingly prized owing to their historical significance as a target of deception. The Sign Verification System (SVS) attempts to establish if a sign is genuine (created by the stated individual) or forged (produced by an impostor). Using images of scanned signatures and other documents without dynamic information about the signing process has proven difficult, especially in offline (static) situations. The use of Deep Learning algorithms to learn feature signature picture representations has been well-documented in the literature over the last five to ten years. Here, we examine how the subject has been studied throughout the last few decades, as well as the most recent developments and future study plans.
DOI:10.1109/ICESC54411.2022.9885550