eprintid: 3740 rev_number: 11 eprint_status: archive userid: 2 dir: disk0/00/00/37/40 datestamp: 2022-10-03 12:49:17 lastmod: 2023-07-17 23:30:13 status_changed: 2022-10-03 12:49:17 type: article metadata_visibility: show creators_name: Sharma, Neha creators_name: Gupta, Sheifali creators_name: Mohamed, Heba G. creators_name: Anand, Divya creators_name: Vidal Mazón, Juan Luis creators_name: Gupta, Deepali creators_name: Goyal, Nitin creators_id: creators_id: creators_id: creators_id: divya.anand@uneatlantico.es creators_id: juanluis.vidal@uneatlantico.es creators_id: creators_id: title: Siamese Convolutional Neural Network-Based Twin Structure Model for Independent Offline Signature Verification ispublished: pub subjects: uneat_eng divisions: uneatlantico_produccion_cientifica divisions: uninimx_produccion_cientifica full_text_status: public keywords: signature verification; two-channel; Siamese network; convolutional neural network; deep learning abstract: One of the toughest biometrics and document forensics problems is confirming a signature’s authenticity and legal identity. A forgery may vary from a genuine signature by specific distortions. Therefore, it is necessary to continuously monitor crucial distinctions between real and forged signatures for secure work and economic growth, but this is particularly difficult in writer-independent tasks. We thus propose an innovative and sustainable writer-independent approach based on a Siamese neural network for offline signature verification. The Siamese network is a twin-like structure with shared weights and parameters. Similar and dissimilar images are exposed to this network, and the Euclidean distances between them are calculated. The distance is reduced for identical signatures, and the distance is increased for different signatures. Three datasets, namely GPDS, BHsig260 Hindi, and BHsig260 Bengali datasets, were tested in this work. The proposed model was analyzed by comparing the results of different parameters such as optimizers, batch size, and the number of epochs on all three datasets. The proposed Siamese neural network outperforms the GPDS synthetic dataset in the English language, with an accuracy of 92%. It also performs well for the Hindi and Bengali datasets while considering skilled forgeries date: 2022-09 publication: Sustainability volume: 14 number: 18 pagerange: 11484 id_number: doi:10.3390/su141811484 refereed: TRUE issn: 2071-1050 official_url: http://doi.org/10.3390/su141811484 access: open language: en citation: Artículo Materias > Ingeniería Universidad Europea del Atlántico > Investigación > Producción Científica Universidad Internacional Iberoamericana México > Investigación > Producción Científica Abierto Inglés One of the toughest biometrics and document forensics problems is confirming a signature’s authenticity and legal identity. A forgery may vary from a genuine signature by specific distortions. Therefore, it is necessary to continuously monitor crucial distinctions between real and forged signatures for secure work and economic growth, but this is particularly difficult in writer-independent tasks. We thus propose an innovative and sustainable writer-independent approach based on a Siamese neural network for offline signature verification. The Siamese network is a twin-like structure with shared weights and parameters. Similar and dissimilar images are exposed to this network, and the Euclidean distances between them are calculated. The distance is reduced for identical signatures, and the distance is increased for different signatures. Three datasets, namely GPDS, BHsig260 Hindi, and BHsig260 Bengali datasets, were tested in this work. The proposed model was analyzed by comparing the results of different parameters such as optimizers, batch size, and the number of epochs on all three datasets. The proposed Siamese neural network outperforms the GPDS synthetic dataset in the English language, with an accuracy of 92%. It also performs well for the Hindi and Bengali datasets while considering skilled forgeries metadata Sharma, Neha; Gupta, Sheifali; Mohamed, Heba G.; Anand, Divya; Vidal Mazón, Juan Luis; Gupta, Deepali y Goyal, Nitin mail SIN ESPECIFICAR, SIN ESPECIFICAR, SIN ESPECIFICAR, divya.anand@uneatlantico.es, juanluis.vidal@uneatlantico.es, SIN ESPECIFICAR, SIN ESPECIFICAR (2022) Siamese Convolutional Neural Network-Based Twin Structure Model for Independent Offline Signature Verification. Sustainability, 14 (18). p. 11484. ISSN 2071-1050 document_url: http://repositorio.unini.edu.mx/id/eprint/3740/1/sustainability-14-11484-v3.pdf