Accepted Unknow ; Published 2024 November 16

Academic Journal of Sociology and Management, 2024, 2(6), 3005-5040.

View PDFDownload PDFArchive (OJS)

This paper presents a novel deep learning-based authentication method for electronic signature verification in financial documents. The proposed system introduces a comprehensive framework integrating YOLOv5-based signature detection, advanced preprocessing techniques, and a Siamese neural network architecture for verification. The system employs a customized feature extraction network incorporating residual connections and attention mechanisms to capture local and global signature characteristics. The implementation includes adaptive preprocessing pipelines and sophisticated loss functions optimized for signature verification tasks. Experimental evaluation on a dataset of 25,000 signature samples from 500 individuals demonstrates superior performance, achieving 98.5% accuracy in verification tasks with a false acceptance rate of 1.2% and a false rejection rate of 1.5%. The system maintains robust performance across various document conditions, demonstrating only a 4.2% accuracy reduction under poor resolution scenarios. Security analysis validates system resilience against adversarial attacks, achieving a 96.5% detection rate. The comprehensive evaluation demonstrates significant improvements over existing accuracy and computational efficiency methods, establishing new benchmarks for signature verification in financial applications. The proposed methodology addresses critical challenges in financial document security while maintaining practical applicability in real-world environments.

Financial Risk Management , Electronic Signature Verification , Deep Learning , Siamese Neural Network , Financial Document Security .

APA Style

Zhang, Y., Bi, W. & Song, R. (2024). Research on deep learning-based authentication methods for e-signature verification in financial documents. Academic Journal of Sociology and Management, 2(6), 35-43. https://doi.org/10.5281/zenodo.14161744

I want to extend my sincere gratitude to Yida Zhu, Keke Yu, Ming Wei, Yanli Pu, and Zeyu Wang for their groundbreaking research on AI-enhanced administrative prosecutorial supervision in financial big data as published in their article titled "AI-Enhanced Administrative Prosecutorial Supervision in Financial Big Data: New Concepts and Functions for the Digital Era"[33]. Their insights and methodologies have significantly influenced my understanding of advanced financial data processing techniques and provided valuable inspiration for my research in signature verification. I would also like to express my heartfelt appreciation to Jiayi Wang, Tianyu Lu, Lin Li, and Decheng Huang for their innovative study on AI-enhanced personalized search approaches, as published in their article titled "Enhancing Personalized Search with AI: A Hybrid Approach Integrating Deep Learning and Cloud Computing"[9]. Their comprehensive analysis of deep learning applications and system architecture design has significantly enhanced my knowledge of neural network implementations and inspired my research in signature verification systems.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.