Nuruzzaman, Mohammad and Hussain, Azham and Mohamad Tahir, Hatim and Abu Seman, Mohamad Amir (2013) Feature extraction using active appearance model algorithm with Bayesian classification approach. In: 4th International Conference on Computing and Informatics (ICOCI 2013), 28 - 30 August, 2013, Sarawak, Malaysia.

PDF Restricted to Registered users only Download (706kB)

Abstract

Face recognition is one of the most important and rapidly advanced active research areas of computer science.In spite of the large number of developed algorithms, real-world performance of face recognition has been disappointing. This study enhances invariant recognition of human faces and analysis to improve face verification and identification performance using Active Appearance Model (AAM) for feature extraction with Bayesian classification approach. This paper addressed some of these issues to bring face recognition more closely to being useful for real-life applications. It directed towards the illumination-invariant automatic recognition of faces and analysis to improve face verification and identification performance.To compare with other feature extraction at the end of the study, an evaluation has been done with an existing face recognition system using AAM algorithm. The experiments performed on part of the FERET color dataset. The result was satisfied with the acceptance rate more than 96%.

Item Type:	Conference or Workshop Item (Paper)
Additional Information:	ISSN 2289-3784 Jointly Organised: Universiti Utara Malaysia, Universiti Malaysia Sabah, College of Art and Sciences, Othman Yeop Abdullah Graduate School of Business
Uncontrolled Keywords:	Face Recognition, Face Detection, Feature Extraction, AAM Algorithm, Bayesian Classification
Subjects:	Q Science > QA Mathematics > QA76 Computer software
Divisions:	College of Arts and Sciences
Depositing User:	Prof. Madya Hatim Mohamad Tahir
Date Deposited:	10 Mar 2014 08:20
Last Modified:	10 Mar 2014 08:20
URI:	https://repo.uum.edu.my/id/eprint/9693

Actions (login required)

View Item