Face Recognition System under Varying Lighting Conditions
Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2013, Vol 14, Issue 3
Abstract
Making recognition more reliable under uncontrolled lighting conditions is one of the most important challenges for practical face recognition systems. Other recognition systems don’t nullify most of the lighting variations. We tackle this by combining the strengths of robust illumination normalization, local texture-based face reresentations, distance transform based matching and kernel based feature extraction and multiple feature fusion. We present a simple and efficient preprocessing chain that eliminates most of the effects of changing illumination while still preserving the essential appearance details that are needed for recognition. We introduce Local Ternary Pattern (LTP), a generalization of the Local Binary Pattern (LBP) local texture descriptor less sensitive to noise. We further increase robustness by introducing Phase Congruency. The resulting method provides a face verification rate of 88.1% at 0.1% false accept rate. Experiments show that our preprocessing method outperforms several existing preprocessors for a range of feature sets, data sets and lighting conditions. We simulate this project using MATLAB software
Authors and Affiliations
P. Kalaiselvi
Keywords
Related Articles
- EP ID EP120689
- DOI -
- Views 91
- Downloads 0
Use of Artificial Neural Networks Type MLP for the Prediction of Phosphorus Level from the Physicochemical Parameters of Sediments
Abstract: The present work is a contribution on the development of mathematical models for predicting the phosphorus contents based on the physicochemical properties of the sediments of the reservoir of the water damSidi...
Dynamic Weighted Query Process in Personalized Privacy Protection in Data Retrieval
Customized web look for (PWS) has demonstrated its efficiency in improving the quality of various look for services on the Internet. However, facts show that users’ desire not to reveal their personal detailsduring look...
An effective citation metadata extraction process based on BibPro parser
There is a dramatic increase in academic publications and these publications are integrated to the digital libraries by making use of citation string. Any author can publish the journals conferences with the help...
A Bayesian Classification Model for Fraud Detection over ATM Platforms
Abstract: The banking system relies greatly on the use of Automated Teller Machines, debit and credit cards as a vital element of its payment processing systems. The efficient functioning of payment processing systems al...
Net Neutrality – A Look at the Future of Internet
Abstract: The Internet is an indispensable medium. The success story of Internet has evolved free from complex government regulation. But there is an increasing threat that Internet Service Providers (ISPs) will al...