Open AccessFacial Expression Identification using Regularized Supervised Distance Preserving Projection
Author(s) -
Sadia Jahan,
Moriyam Akter,
Sifta Yeasmin,
Farhana Ahmed Simi
Publication year - 2021
Publication title -
the dhaka university journal of science
Language(s) - English
Resource type - Journals
eISSN - 2408-8528
pISSN - 1022-2502
DOI - 10.3329/dujs.v69i2.56485
Subject(s) - artificial intelligence , computer science , classifier (uml) , pattern recognition (psychology) , projection (relational algebra) , facial recognition system , feature extraction , facial expression recognition , facial expression , expression (computer science) , feature (linguistics) , computer vision , algorithm , linguistics , philosophy , programming language
Facial expression recognition is one of the most reliable and a key technology of advanced human-computer interaction with the rapid development of computer vision and artificial intelligence. Nowadays, there has been a growing interest in improving expression recognition techniques. In most of the cases, automatic recognition system’s efficiency depends on the represented facial expression feature. Even the best classifier may fail to achieve a good recognition rate if inadequate features are provided. Therefore, feature extraction is a crucial step of the facial expression recognition process. In this paper, we have used Regularized Supervised Distance Preserving Projection for extracting the best features of the images. Numerical experiment shows that the use of this technique outperforms many of state of art approaches in terms of recognition rate.Dhaka Univ. J. Sci. 69(2): 70-75, 2021 (July)