Open AccessA Gender Recognition System Using Facial Images with High Dimensional Data
Author(s) -
Martins E. Irhebhude,
Adeola O. Kolawole,
Hauwa K. Goma
Publication year - 2021
Publication title -
malaysian journal of applied sciences
Language(s) - English
Resource type - Journals
ISSN - 0127-9246
DOI - 10.37231/myjas.2021.6.1.275
Subject(s) - artificial intelligence , support vector machine , pattern recognition (psychology) , local binary patterns , computer science , classifier (uml) , histogram , histogram of oriented gradients , facial recognition system , feature extraction , computer vision , image (mathematics)
Gender recognition has been seen as an interesting research area that plays important roles in many fields of study. Studies from MIT and Microsoft clearly showed that the female gender was poorly recognized especially among dark-skinned nationals. The focus of this paper is to present a technique that categorise gender among dark-skinned people. The classification was done using SVM on sets of images gathered locally and publicly. Analysis includes; face detection using Viola-Jones algorithm, extraction of Histogram of Oriented Gradient and Rotation Invariant LBP (RILBP) features and trained with SVM classifier. PCA was performed on both the HOG and RILBP descriptors to extract high dimensional features. Various success rates were recorded, however, PCA on RILBP performed best with an accuracy of 99.6% and 99.8% respectively on the public and local datasets. This system will be of immense benefit in application areas like social interaction and targeted advertisement.