EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Classification of Offline Signature Authenticity Based on Digital Imagery Using Convolutional Neural Network

Sartika Lina Mulani Sitio, Sunardi Sunardi, Abdul Fadlil

Computer Journal · 01.02.2027 · 5 (1) · 142-151

DOI: 10.58477/cj.v5i1.608 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer Journal, online erschienen am 01.02.2027. Verfasst von Sartika Lina Mulani Sitio und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Verification of digital image-based signature authenticity is an automated approach to distinguishing between genuine and forged signatures. This study aims to develop a classification model for signature authenticity using a Convolutional Neural Network (CNN). The study used the CEDAR dataset, which consists of 2,640 images comprising 1,320 genuine signatures and 1,320 forged signatures from 55 writers. The research stages included image preprocessing through resizing to 128 × 128 pixels and pixel value normalization, stratified dataset splitting, data augmentation on the training subset, and training of a CNN consisting of four convolutional blocks with regularization. Model evaluation was conducted on 396 test images using accuracy, precision, recall, F1-score, a confusion matrix, and the Receiver Operating Characteristic (ROC) curve. The test results showed that the model achieved an accuracy of 98.48%, with 390 correctly classified images and six misclassified images. These findings indicate that the developed CNN model can classify genuine and forged signatures with high accuracy under the dataset and testing scheme applied. However, these results do not establish the superiority of the proposed architecture over other methods because this study did not include a comparative model.

Zitieren

APA

Sitio, S. L. M., Sunardi, S., & Fadlil, A. (2027). Classification of Offline Signature Authenticity Based on Digital Imagery Using Convolutional Neural Network. Computer Journal, 5(1), 142-151. https://doi.org/10.58477/cj.v5i1.608

Vancouver

Sitio SLM, Sunardi S, Fadlil A. Classification of Offline Signature Authenticity Based on Digital Imagery Using Convolutional Neural Network. Computer Journal. 2027;5(1):142-151. doi:10.58477/cj.v5i1.608

BibTeX
@article{Sitio2027,
  author = {Sitio, Sartika Lina Mulani and Sunardi, Sunardi and Fadlil, Abdul},
  title = {Classification of Offline Signature Authenticity Based on Digital Imagery Using Convolutional Neural Network},
  journal = {Computer Journal},
  year = {2027},
  volume = {5},
  number = {1},
  pages = {142--151},
  doi = {10.58477/cj.v5i1.608}
}
Kostenlos starten →