On the Neural Networks for Biometric Authentication Based on Keystroke Dynamics
| Autor: | LEE, KEN-YU, 李耕瑜 |
|---|---|
| Rok vydání: | 2017 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 105 In this study, we propose a biometric authentication method to identify and block illegal users, even if the whole password is exposed. Our method simultaneously records scan codes and keystroke sequence of passwords; furthermore, by deep learning of convolutional neural networks, legal users can be effectively distin- guished from illegal users. The experimental results show that illegal users are successfully blocked even if the password has been exposed. Although the aver- age login failure rate of legal users is 6 percent, they can reenter passwords once to be admitted. We also compare recognition rates between convolutional neural networks and neural networks and prove that convolu- tional neural networks are better. Finally, by GPU parallel computing, we further obtain about 5 times acceleration of system performance. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|