Call For Paper Volume:4 Issue:8 Aug'2017 |

Face Recognition Using Singular Value Decomposition and Hidden Markov Model

Publication Date : 01/11/2015

Author(s) :

Memane Swati Bhaskar , Prof. Dr. P.D.Bhamre.

Volume/Issue :
Volume 2
Issue 10
(11 - 2015)

Abstract :

Biometrics is an automatic method of recognizing a person based on his physiological characteristic. Face recognition system (FRS) is an emerging field in biometrics. FRS can be used for identifying a person in highly secured areas as well as where person authentication is required; it is one of the best methods because face structure cannot be copied like a password. In most of the face recognition system front pose face images are considered when the database is created. But when the query image is not its front face image but side view of the face or any other pose or facial expression other than the image in the database, the system fails to identify the person. In this paper a method to solve this problem is given. The database contains the facial images of different persons with different facial expressions and poses. The features are extracted using Singular value decomposition method (SVD). A HMM (Hidden Markov Model) is used to train these features of the images from the given set of data. These resultant HMM’s then compared with the HMM of the available face image to identify the person. This method gives good results compared to other 2D face recognition systems which are affected by pose variations and intensity of the 2D image. The system is tested using ORL standard database and the algorithm for this system is simulated using MATLAB software.

No. of Downloads :



Web Design MymensinghPremium WordPress ThemesWeb Development

Face Recognition Using Singular Value Decomposition and Hidden Markov Model

October 31, 2015