DocumentCode :
31282
Title :
Interactive Artificial Bee Colony Supported Passive Continuous Authentication System
Author :
Pei-Wei Tsai ; Khan, Muhammad Khurram ; Jeng-Shyang Pan ; Bin-Yih Liao
Author_Institution :
Dept. of Electron. Eng., Nat. Kaohsiung Univ. of Appl. Sci., Kaohsiung, Taiwan
Volume :
8
Issue :
2
fYear :
2014
fDate :
Jun-14
Firstpage :
395
Lastpage :
405
Abstract :
In this paper, an initiative passive continuous authentication (CA) system based on both hard and soft biometrics is presented. Human facial features are used as hard biometric information for the authentication process, and the clothes´ color of a user is employed as the soft biometric information. The passive CA system keeps verifying, without interrupting the user from concentrating on his work. It also provides the capacity for the machine to recognize who is in front of the terminal, reduces the potential security leaks, and denies access to the invader with the stolen account and password. In this system, the face recognition core is implemented not only by the Eigenface method, but also assisted by the interactive artificial bee colony optimization algorithm. The proposed method is evaluated by the ORL face database and tested on the prototype CA system for computer security. The experimental results indicate that the accuracy of recognition is raised up to 3.13%, i.e., from 83.75% to 86.88%, with data from the ORL database, and it is improved by 34.53% on average in the real-time continuous authentication environment.
Keywords :
face recognition; optimisation; Eigenface method; ORL face database; face recognition core; hard biometric information; human facial features; interactive artificial bee colony optimization algorithm; passive CA system; passive continuous authentication system; real-time continuous authentication environment; Authentication; Biometrics; Face; Face recognition; Image color analysis; Skin; Vectors; Artificial bee colony (ABC); biometric; continuous authentication (CA); face recognition; passive authentication;
fLanguage :
English
Journal_Title :
Systems Journal, IEEE
Publisher :
ieee
ISSN :
1932-8184
Type :
jour
DOI :
10.1109/JSYST.2012.2208153
Filename :
6264083
Link To Document :
بازگشت