|Title:||Age invariant face recognition|
|Advisors:||Lam, K. M. Kenneth (EIE)|
|Subject:||Hong Kong Polytechnic University -- Dissertations|
Human face recognition (Computer science)
Image analysis -- Data processing
Image processing -- Digital techniques
|Department:||Department of Electronic and Information Engineering|
|Pages:||45 pages : color illustrations|
|Abstract:||In this dissertation, different algorithms have been investigated for age-invariant face recognition. The first step is to study and evaluate existing face recognition methods on face images with different ages. Then based on an age-invariant face recognition method, named Multi-Feature Discriminant Analysis (MFDA), further modifications are applied to this method in order to improve the recognition rate on face images with age progression. The modifications include replacing the feature extraction methods with other effective algorithms, experimenting different similarity measurements, and score fusion methods. Since Local Binary Pattern has many variants, which are created to deal with different conditions, some of these variants are selected to replace the original Local Binary Pattern in the MFDA method. For example, the Multi-scale Local Binary Pattern is replaced by Local Ternary Pattern and Cosine similarity is also tested as the feature distance measurement. Experiments were conducted and the results shows that with the FG-NET database, replacing the distance measurement method leads to the greatest improvement in terms of recognition rate.|
|Rights:||All rights reserved|
Files in This Item:
|991022131146203411.pdf||For All Users (off-campus access for PolyU Staff & Students only)||1.04 MB||Adobe PDF||View/Open|
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item: