|
Abstract
|
Article Information:
Performance Evaluation of LDA, CCA and AAM
M. Jasmine Pemeena Priyadarsini, K. Murugesan, Srinivasa Rao Inbathini, J. Vishal, S. Anand and Rahul N. Nair
Corresponding Author: M. Jasmine Pemeena Priyadarsini
Submitted: July 18, 2014
Accepted: October 17, 2014
Published: March 25, 2015 |
Abstract:
|
Wouldn't we love to replace passwords access control to avoid theft, forgotten passwords? Wouldn't we like to enter the security areas just in seconds? Yes the answer is face recognition. In this study we explore and compare the performance of three algorithms namely LDA, CCA, AAM. LDA (an evolution of PCA is a dimensionality reduction technique where it solves the problem of illumination to some extent, maximizing the inter class separation and minimizing the intra class variations. CCA, a measure of linear relationship between two multidimensional variables where it takes the advantage of PCA and LDA for maximizing the correlation and better performance. AAM is a model based approach where it just picks the landmarks of the images for recognition therefore reducing the error rate and producing good performance rate.
Key words: AAM, CCA, efficiency, face recognition, landmarks, LDA, PCA, performance
|
|
Abstract
|
PDF
|
HTML |
|
Cite this Reference:
M. Jasmine Pemeena Priyadarsini, K. Murugesan, Srinivasa Rao Inbathini, J. Vishal, S. Anand and Rahul N. Nair, . Performance Evaluation of LDA, CCA and AAM. Research Journal of Applied Sciences, Engineering and Technology, (9): 685-699.
|
|
|
|
 |
ISSN (Online): 2040-7467
ISSN (Print): 2040-7459 |
 |
| Information |
|
|
|
|
| Sales & Services |
|
|
|
|
