Face Detection and Recognition for Security System using Local Binary Patterns (LBP)

B.A.S.S. Karunarathne (Department of Computing and Information Systems, Wayamba University of Sri Lanka, Kuliyapitiya, Sri Lanka)
W. H. C. Wickramaarachchi (Department of Computing and Information Systems, Wayamba University of Sri Lanka, Kuliyapitiya, Sri Lanka)
K.K.K.M.C. De Silva (Department of Computing and Information Systems, Wayamba University of Sri Lanka, Kuliyapitiya, Sri Lanka)

Journal of ICT, Design, Engineering and Technological Science
Volume 3, Issue 1

Facial recognition has been used in several areas such as security, detection of criminals or suspects, etc. The face is considered the most important part of the human body in the person identification process since it can be considered the most important perspective in human identification. This paper proposed a real-time monitoring security system for high secured areas rather than relying on human security officers. The proposed system engages with capturing and detecting the human face using the real-time image segment captured by the web camera through the Viola-Jones algorithm. After detecting the face in the captured image, face identification can be made through the LBP by finding the similarities. Once the identification is made, automatically it recognizes the face. Open Computer Vision (OpenCV) is one of the most efficient machine learning libraries used for Object Detection. OpenCV library is used for detection, and Haar - Cascades are trained to detect certain objects. LBP is used for the face recognition process.