An Image-Based Method of Distinguishing Children from Adults

Authors

  • Alireza Keshavarz Choobeh Author

Keywords:

Age estimation, active appearance model, LDA, minimum distance classifier.

Abstract

Distinguishing children from adults via facial 
image analysis has lots of potential real-world applications such 
as security access control and human computer interaction. 
However, it is still a challenging problem for the computer 
vision systems to automatically and effectively distinguish 
children from adults. In this paper, we introduce a novel 
children recognition method, which improves both the accuracy 
and reliability of the latest work on this subject simultaneously. 
Results on the FG-NET aging database show that using the 
minimum distance classifier on the one dimensional feature 
space created by using Active Appearance Model (AAM) 
followed by Linear Discriminant Analysis (LDA), we can 
recognize the children and adults with accuracies up to 89% 
and 90%, respectively.

Downloads

Download data is not yet available.

Downloads

Published

19.09.2013

How to Cite

An Image-Based Method of Distinguishing Children from Adults. (2013). International Journal of Information and Electronics Engineering, 3(5), 533-535. https://www.ijiee.org/index.php/ijiee/article/view/740