Scholarlink

<p>AI-Synergistic effect refers to the enhancement of the outcome achieved through the combined integration of diverse components and/or Artificial Intelligence algorithms, surpassing the sum of their individual contributions. Online Signature-Trace-Flow Profiling is emerging as a distinct online signature recognition approach, delineating itself from Offline Signature Recognition across various aspects. Its applicability spans diverse online authentication scenarios, including web and mobile applications, computers, and server access, as well as the verification of online documents. While digital signatures serve a similar purpose in authenticating and securing electronic documents, their widespread adoption among the general population remains limited. Currently, conventional handwritten signatures remain widely accepted for document handling. Addressing this landscape, collaborative efforts focus on seamlessly integrating online application Sign-In and document authentication through an easily accessible web/mobile application. Online Signature-Flow Profiling is introduced to meet the dual requirements of authentication and verification. The proposed method comprises three core modules: the Signature-Trace-Flow Registration Module, the Signing Module, and the Receiver Verification Module. The Signature-Trace-Flow Registration Module captures and registers the flow of signature traces through online signing. To enhance verification accuracy, a novel Optimized-Spherical-CRNN algorithm is proposed, tasked with matching patterns in the registered traces against online recordings. This optimized spherical CRNN leverages 3D signature point cloud processing via spherical transformation. The Average Equal Error Rate (% AEER) is notably lower compared to alternative methods, contributing to heightened recognition system accuracy. Online Signature-Trace-Flow Profiling and Hash-QR Encoding algorithms are combined and utilized to forge signature detection, document protection and various applications authentications. False recognized signatures with visible similarity are detected significantly using the proposed algorithm.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

AI-synergistic effect in C-RNN based online signature profiling and hash-QR encoding for forge signature detection, documents protection and applications authentication

  • 期刊论文

  • 发表日期：

    2025年7月2日

查看全文
  • Hemant Wani,
  • Kantilal P. Rane,
  • Vijay M. Deshmukh,
  • Kushal D. Badgujar,
  • J. Jeyvel,
  • Somnath Wategaonkar

摘要

AI-Synergistic effect refers to the enhancement of the outcome achieved through the combined integration of diverse components and/or Artificial Intelligence algorithms, surpassing the sum of their individual contributions. Online Signature-Trace-Flow Profiling is emerging as a distinct online signature recognition approach, delineating itself from Offline Signature Recognition across various aspects. Its applicability spans diverse online authentication scenarios, including web and mobile applications, computers, and server access, as well as the verification of online documents. While digital signatures serve a similar purpose in authenticating and securing electronic documents, their widespread adoption among the general population remains limited. Currently, conventional handwritten signatures remain widely accepted for document handling. Addressing this landscape, collaborative efforts focus on seamlessly integrating online application Sign-In and document authentication through an easily accessible web/mobile application. Online Signature-Flow Profiling is introduced to meet the dual requirements of authentication and verification. The proposed method comprises three core modules: the Signature-Trace-Flow Registration Module, the Signing Module, and the Receiver Verification Module. The Signature-Trace-Flow Registration Module captures and registers the flow of signature traces through online signing. To enhance verification accuracy, a novel Optimized-Spherical-CRNN algorithm is proposed, tasked with matching patterns in the registered traces against online recordings. This optimized spherical CRNN leverages 3D signature point cloud processing via spherical transformation. The Average Equal Error Rate (% AEER) is notably lower compared to alternative methods, contributing to heightened recognition system accuracy. Online Signature-Trace-Flow Profiling and Hash-QR Encoding algorithms are combined and utilized to forge signature detection, document protection and various applications authentications. False recognized signatures with visible similarity are detected significantly using the proposed algorithm.