Handwritten Signature Validation and Counterfeit Detection Framework with Knn, Backpropagation, and Convolutional Neural Networks

Authors

Mr Jadhav U B, Prof. Kharat Y. D, Prof. Tirmakhe V. R, Prof. Bhmabare A. V, Prof. Bansode D. K

Abstract

Biometrics has gained global prominence as a reliable method for identifying and verifying individuals, including their handwritten signatures. A person's handwritten signature is a distinct and personal identifier, widely utilized in banking, financial, and legal operations. However, handwritten signatures have become increasingly susceptible to forgery due to their historical and legal significance. The Signature Verification System (SVS) aims to determine the authenticity of a signature, identifying whether it is genuine (created by the claimed individual) or forged (produced by an imposter).
Verifying signatures using static images of scanned signatures, without dynamic information about the signing process (such as speed and pressure), poses significant challenges, particularly in offline scenarios. Over the past decade, deep learning algorithms have demonstrated their effectiveness in extracting and learning the unique features of signature images. This field of research has seen continuous progress, and recent developments provide insight into how signature verification has evolved, along with promising directions for future exploration.

Keywords

Signature Verification, KNN, CNN, Backproagation.

Handwritten Signature Validation and Counterfeit Detection Framework with Knn, Backpropagation, and Convolutional Neural Networks. Mr Jadhav U B, Prof. Kharat Y. D, Prof. Tirmakhe V. R, Prof. Bhmabare A. V, Prof. Bansode D. K. 2024. IJIRCT, Volume 10, Issue 5. Pages 1-5. https://www.ijirct.org/viewPaper.php?paperId=2410048