Real Time Face Recognition Using LBP Features

摘要

Facial identication is important these days, Methods are required must be fast and accurate enough to work realtime. So research in methods for face recognition seems ever growing. The measurements of an individuals data is inherent part of FR techniques. In biomedical verification and identification requires the dataset such as iris, finger prints etc. While for FR, cameras are replacing the cards at places like ATMs. It helps capturing facial images of customers, and compare these photos to the images of account holders database of banks to verify the customers identity. This paper proposes way for face recognition. Shape and texture of facial images are studied for representation. The face is Firstly divided into tiny regions. Which are used to get LBPH. These histograms merged in one partially enhanced histogram, by which we get face images effciently. KNN classier does the classication and not just effciency but the simplicity of manner allows very fast feature extraction.