We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.CV

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computer Vision and Pattern Recognition

Title: Multimodal Personal Ear Authentication Using Smartphones

Abstract: In recent years, biometric authentication technology for smartphones has become widespread, with the mainstream methods being fingerprint authentication and face recognition. However, fingerprint authentication cannot be used when hands are wet, and face recognition cannot be used when a person is wearing a mask. Therefore, we examine a personal authentication system using the pinna as a new approach for biometric authentication on smartphones. Authentication systems based on the acoustic transfer function of the pinna (PRTF: Pinna Related Transfer Function) have been investigated. However, the authentication accuracy decreases due to the positional fluctuation across each measurement. In this paper, we propose multimodal personal authentication on smartphones using PRTF. The pinna image and positional sensor information are used with the PRTF, and the effectiveness of the authentication method is examined. We demonstrate that the proposed authentication system can compensate for the positional changes in each measurement and improve robustness.
Comments: 9 pages, 23 figures
Subjects: Computer Vision and Pattern Recognition (cs.CV); Cryptography and Security (cs.CR); Image and Video Processing (eess.IV)
Cite as: arXiv:2103.12575 [cs.CV]
  (or arXiv:2103.12575v1 [cs.CV] for this version)

Submission history

From: Shunji Itani [view email]
[v1] Tue, 23 Mar 2021 14:19:15 GMT (2560kb)

Link back to: arXiv, form interface, contact.