保存先:
| 第一著者: | |
|---|---|
| フォーマット: | Recurso digital |
| 言語: | |
| 出版事項: |
Zenodo
2024
|
| 主題: | |
| オンライン･アクセス: | https://doi.org/10.5281/zenodo.18172514 |
| タグ: |
タグ追加
タグなし, このレコードへの初めてのタグを付けませんか!
|
目次:
- In practical applications, unimodal biometric systems rely on evidence from a single source of information for authentication. However, these systems encounter challenges, including noise in the sensed data arising from factors such as the repeated use of fingerprint sensors and intra-class variation. For instance, the reliability of a fingerprint-based system may be compromised due to variations in the sensed data caused by repeated sensor use. To address these limitations, the adoption of a multi-modal biometric authentication system becomes pivotal. This approach enhances the robustness and reliability of the authentication process by incorporating multiple biometric traits. The utilization of multiple modalities ensures that the system is better equipped to discern the presence of a live user at the data acquisition point, thereby mitigating the impact of potential challenges associated with unimodal systems. Common examples of multimodal biometrics involve the fusion of different biometric modalities, such as face and fingerprint or iris and fingerprint. By integrating diverse biometric sources, a multimodal approach leverages the strengths of each modality, contributing to a more resilient and effective authentication system in real-world scenarios.