19-02-2018, 03:00 PM
Abstract
In this letter, we present a novel biometric authentication system to identify the identities of a person by our patio print. Unlike existing paypalprint systems for criminal applications, the proposed system is targeted at civil applications, which in real time requires a person to identify a large database with high confidence. The system is built by four major components: User Interface module, Acquisition module, Recognition module and external module. More than 7,000 tree-shaped images have been collected to test the system's performance. The system can identify less than 400 palms with false acceptance rate, 0.02%, and high actual acceptance rate, 98.83%. For verification, the system can work on false acceptance rate, 0.017% and a false rejection rate, 0.86%. The execution time is less than 1 second for the entire process including image collection, preprocessing, feature extraction and matching.
In this letter, we present a novel biometric authentication system to identify the identities of a person by our patio print. Unlike existing paypalprint systems for criminal applications, the proposed system is targeted at civil applications, which in real time requires a person to identify a large database with high confidence. The system is built by four major components: User Interface module, Acquisition module, Recognition module and external module. More than 7,000 tree-shaped images have been collected to test the system's performance. The system can identify less than 400 palms with false acceptance rate, 0.02%, and high actual acceptance rate, 98.83%. For verification, the system can work on false acceptance rate, 0.017% and a false rejection rate, 0.86%. The execution time is less than 1 second for the entire process including image collection, preprocessing, feature extraction and matching.