Study of Different Algorithms for Face Recognition

Prakash, A and Tewari, M K (2010) Study of Different Algorithms for Face Recognition. BTech thesis.



The importance of utilising biometrics to establish personal authenticity and to detect impostors is growing in the present scenario of global security concern. Development of a biometric system for personal identification, which fulfils the requirements for access control of secured areas and other applications like identity validation for social welfare, crime detection, ATM access, computer security, etc., is felt to be the need of the day [2]. Face recognition has been evolving as a convenient biometric mode for human authentication for more than last two decades. It plays an important role in applications such as video surveillance, human computer interface, and face image database management [1]. A lot of techniques have been applied for different applications. Robustness and reliability becomes more and more important for these applications especially in security systems.

Basically Face Recognition is the process through which a person is identified by his facial image. With the help of this technique it is possible to use the facial image of a person to authenticate him into any secure system. Face recognition approaches for still images can be broadly categorized into holistic methods and feature based methods. Holistic methods use the entire raw face image as an input, whereas feature based methods extract local facial features and use their geometric and appearance properties.

This work studies the different approaches for a Face Recognition System. The different approaches like PCA, DCT and different types of Wavelets have been studied with the help of Euclidean distance as a classifier and Neural Network as a classifier.

The results have been compared for the two database, AMP which contains 975 images of 13 individuals (each person has 75 different images) under various facial expressions and lightning condition with each image being cropped and resized to 64×64 pixels for the simulation and ORL (Olivetti Research Lab) which contains 400 images (each with 112×92 pixels) corresponding to 40 persons in 10 poses each including both male and female. The ORL database image has been resized to 128×128 pixels.

Item Type:Thesis (BTech)
Uncontrolled Keywords:Face Recognition
Subjects:Engineering and Technology > Electronics and Communication Engineering > Image Processing
Divisions: Engineering and Technology > Department of Electronics and Communication Engineering
ID Code:1701
Deposited By:Anshuman Prakash
Deposited On:13 May 2010 15:14
Last Modified:13 May 2010 15:14
Related URLs:
Supervisor(s):Meher, S

Repository Staff Only: item control page