Palm-print recognition based on deep residual networks | ||||
| International Journal of Telecommunications | ||||
| Volume 04, Issue 02, July 2024, Page 1-18 PDF (2.2 MB) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/ijt.2024.333452.1066 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
tarek awad elbendary  1; Abeer Khalil2; mahmoud M. Saafan3; Hossam Moustafa 4
| ||||
| 1engineer.tarekawadelbendary@std.mans.edu.eg; M.Sc. student at the Department of Electronics and Communications Engineering at the Faculty of Engineering, Mansoura University 1 | ||||
| 2Associate Professor at the Department of Electronics and Communications Engineering, at the Faculty of Engineering, Mansoura University | ||||
| 3Associate Professor at the Department of Computers Engineering and Control systems, at the Faculty of Engineering, Mansoura University | ||||
| 4Professor at the Department of Electronics and Communications Engineering at the Faculty of En-gineering, Mansoura University | ||||
| Abstract | ||||
| A palmprint is a tiny portion of the palm flat that carries extra information that may be utilized in authentication systems. It also has the quality of permanence, meaning that it will not change throughout time. Extracting meaningful characteristics from palm prints. PPR offers high precise accuracy in authentication operations, especially in civilian applications and military and law enforcement applications. Civilian applications play vital roles in sectors such as biometric identification, forensic investigations, banking and financial services, time and attendance systems, and military and law enforcement applications such as border control and immigration, military access control, criminal databases, and counterterrorism operations. The majority of newly developed approaches rely on primary lines, wrinkles, and creases, which are insufficient to discriminate between two people owing to their proximity. Deep learning approaches are increasingly used for extracting deep properties such as texture. We introduce a deep residual neural network (RESENT) built for safe authentication using palmprint photos. Experiments were conducted using the CASIA, IIT Delhi Touchless, and SMPD Palm-Print databases, with accuracy and F1-score employed for assessment. The suggested model was highly accurate, scoring 99.75 percent. This approach for palmprint authentication is efficient and effective. | ||||
| Keywords | ||||
| PPR; biometric; DL; convolutional neural network; authentication | ||||
|
Statistics Article View: 187 PDF Download: 223 |
||||
