Age Classification using Multiple Deep Convolutional Neural Network

  • Khaled Rahman Hassan, Israa Hadi Ali


With the giant leap in social media and multimedia systems, the age classification from humans face images has become one of the most crucial active research topics during the past years.Nevertheless, there are still problems with age classification and age estimation systems in real-world applications. In this work, we suggest an approach to age classification using multiple convolutional neural networks (CNN) and based on ensemble methods. Our proposed system involves five sequential phases as follows: face detection, remove background, face alignment, multiple CNN, and Voting Ensemble. The various CNN network consists of three different sub-CNN in structure and depth; the goal of this difference It is to extract different features for each network. Each network is trained separately on the AGFW dataset, and then we use Voting Ensemble to combine predictions. Finally, we evaluated the performance of the three sub-CNN and compared them with the proposed system.

Keywords: age classification,age estimate, convolutional neural networks(CNN),computer vision.

How to Cite
Khaled Rahman Hassan, Israa Hadi Ali. (2020). Age Classification using Multiple Deep Convolutional Neural Network. International Journal of Advanced Science and Technology, 29(04), 5518 - 5527. Retrieved from