Identificador persistente para citar o vincular este elemento: http://hdl.handle.net/10553/133375
Título: Janus-Faced Handwritten Signature Attack: A Clash Between a Handwritten Signature Duplicator and a Writer Independent, Metric Meta-learning Offline Signature Verifier
Autores/as: Giazitzis, Alexios
Diaz, Moises 
Zois, Elias N.
Ferrer, Miguel A. 
Clasificación UNESCO: 33 Ciencias tecnológicas
Palabras clave: Offline Signature Verification
Performance Evaluation
Riemannian Manifold
Sigmml
Synthetic Signature Attack
Fecha de publicación: 2024
Conferencia: 18th International Conference on Document Analysis and Recognition (ICDAR 2024)
Resumen: Signature verification is a popular research area. SigmML, a new system for offline, writer-independent verification, has been developed, offering a unique approach outside typical Euclidean network learning methods. This verifier operates in the space of symmetric positive definite matrices and has demonstrated promising preliminary state-of-the-art results in intra and cross lingual dataset experiments. However, any offline automatic signature verifier faces a potential vulnerability: susceptibility to massive attacks using synthetic signatures. This concern becomes more pronounced given the significant advancements in handwritten image generation techniques. To evaluate the threat level of synthetic attacks to the original version of SigmML, we assess its performance under several attack profiles involving the duplication of synthetically questioned signatures, which are used during the test stage. These profiles advance the threat level to the SigmML verifier by refining the output of the duplicator with a quality control mechanism which intuitively adapts the a-priori knowledge of the intra-variability of each writer. In our experiments, we considered signatures written in various countries and styles, including specimens in Western, Devanagari, and Bengali scripts. Quantitatively, we demonstrate this delicate security issue in the context of signature verification. The proposed attack profiles significantly degrade the performance of SigmML, surpassing the results obtained against skilled forgery experiments by more than double.
URI: http://hdl.handle.net/10553/133375
ISBN: 9783031705359
ISSN: 0302-9743
DOI: 10.1007/978-3-031-70536-6_13
Fuente: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)[ISSN 0302-9743],v. 14805 LNCS, p. 216-232, (Enero 2024)
Colección:Actas de congresos
Vista completa

Google ScholarTM

Verifica

Altmetric


Comparte



Exporta metadatos



Los elementos en ULPGC accedaCRIS están protegidos por derechos de autor con todos los derechos reservados, a menos que se indique lo contrario.