Improved digital authentication systems that allow users to enroll with one finger and unlock with any other finger.
Background:
Fingerprints have long been regarded as one of the most reliable forms of biometric identification due to their presumed uniqueness. The distinct patterns and ridges on each individual's fingertips have served as a cornerstone in forensic science and digital authentication systems. However, recent research has cast doubt on the absolute uniqueness of fingerprints, particularly within the same individual.
Technology Overview:
This University at Buffalo and Columbia University technology shows the correlation between fingerprints from different fingers of the same person, challenging conventional assumptions about fingerprint uniqueness. By utilizing deep learning models to analyze high-quality fingerprint images, the research aims to improve efficiency and accuracy in digital authentication and forensic investigations.
https://buffalo.technologypublisher.com/files/sites/7570_in-part_image.jpg
Source: Kitreel, https://stock.adobe.com/uk/145560454, stock.adobe.com
Advantages:
Applications:
Intellectual Property Summary:
Provisional patent application 63/516,302 filed July 28, 2023.
Stage of Development:
TRL2
Licensing Status:
Available for licensing or collaboration.
Figure 1:
https://buffalo.technologypublisher.com/files/sites/7570_image.jpg