Development of a real-time intelligent biometric face detection and recognition system in LabVIEW

Muralindran Mariappan and Manimehala Nadarajan and Rosalyn R. Porle (2013) Development of a real-time intelligent biometric face detection and recognition system in LabVIEW. American Journal of Intelligent Systems, 3 (1). pp. 40-49. ISSN 2165-8994


Download (125kB) | Preview


Face detection and recognition plays a vital role with broad application in areas like crowd surveillance, security system, human computer interface, etc. In principle, biometric system is preferred for people identification due to its reliability and accuracy. The biggest challenge in face recognition arises when a real-time application system is designed for frontal and non-frontal images. The variations in face poses and expressions greatly impact the identification accuracy of a moving person. To circumvent this issue, in this paper, a real-time biometric system using face region is designed to detect and recognize a person in a pre-defined range using LabVIEW. Face region is proposed to eliminate any physical contact with the system. Neural Network (NN) is employed by training the face images in different distance and angle which allows this system to work for frontal and non- frontal face recognition. Algorithms in LabVIEW are developed to detect and extract the face region in a captured frame which is then sent to NN for recognition process. Consecutive frames video processing was implemented for a real-time face recognition system. About 128 images were used for training and 160 images were tested and it achieves an accuracy of 96.8% in real-time testing.

Item Type: Article
Uncontrolled Keywords: Face Detection, Face Recognition, Frontal Images, Non- frontal Images, Neural Network, LabVIEW
Subjects: T Technology > TJ Mechanical engineering and machinery
Divisions: FACULTY > Faculty of Engineering
Depositing User: ADMIN ADMIN
Date Deposited: 16 Dec 2016 10:04
Last Modified: 23 Oct 2017 16:42

Actions (login required)

View Item View Item