Real-Time Human Face Recognition Using Local Binary Patterns and Principle Component Analysis

International Journal of Computer Science and Technology | Vol. 11, No. 8, August 2009 | pp. 57–64

DOI: 10.46882/2009/IJCST/000017

Article Type: Research Article

Title: Real-Time Human Face Recognition Using Local Binary Patterns and Principle Component Analysis

Names of Authors: Fatima Al-Sayed¹, Youssef Mansour²

Authors’ Affiliations: ¹Department of Computer Engineering, Cairo University, Giza, Egypt; ²Faculty of Information Technology, King Abdullah University of Science and Technology, Thuwal, Saudi Arabia

Abstract: Computer vision systems deployed in surveillance frameworks require high facial recognition accuracy alongside computational lightness for real-time video processing. This paper introduces an integrated face recognition framework combining Local Binary Patterns (LBP) for robust texture feature extraction with Principal Component Analysis (PCA) for dimensional reduction. By applying LBP prior to projection, our method maintains resilience against variations in ambient illumination and subtle facial expressions. Testing was executed using the extended Yale Face Database B and the ORL database. The hybrid LBP-PCA system achieved a peak recognition accuracy of 97.8% under variable lighting environments, outperforming standard Eigenfaces approaches by 6.4%. Execution latency tests registered an average matching time of 15.1 milliseconds per frame on standard hardware configurations, rendering it suitable for real-time security installations. The results prove that localized spatial descriptor fusion significantly strengthens global projection matrices against environmental noise without elevating computational costs.

Keywords: Face Recognition, Local Binary Patterns, Principal Component Analysis, Feature Extraction, Computer Vision, Surveillance Systems

Manuscript Timeline: Received: January 11, 2009; Revised: February 22, 2009; Accepted: March 15, 2009; Published: August 14, 2009

Citation: Al-Sayed, F., & Mansour, Y. (2009). Real-Time Human Face Recognition Using Local Binary Patterns and Principle Component Analysis. International Journal of Computer Science and Technology, 11(8), 57–64.

Subscribe to read the full article: https://www.internationalscholarsjournals.org/subscribe-to-read