Analyzing the Importance of Age-biased Data in Recognizing Emotions from Facial Expressions Using Custom Data Sets and CNN Algorithm

Authors

  • Hyungjoo Park Research Center for Advanced Convergence Technology, Korea Electronics-Machinery Convergence Technology Institute, Korea.
  • Youngha Shin Research Center for Advanced Convergence Technology, Korea Electronics-Machinery Convergence Technology Institute, Korea.
  • Kyu Song Research Center for Advanced Convergence Technology, Korea Electronics-Machinery Convergence Technology Institute, Korea.
  • Channyeong Yun Research Center for Advanced Convergence Technology, Korea Electronics-Machinery Convergence Technology Institute, Korea.
  • Dongyoung Jang Korea Electronics-Machinery Convergence Technology Institute, Seoul National University of Science and Technology, Korea.

DOI:

https://doi.org/10.9734/bpi/rpst/v9/5026B

Keywords:

Facial emotion recognition, deep learning, onvolution neural networks, age bias

Abstract

This paper explores the importance of age-biased data in recognizing emotions from facial expressions. A custom data set was created by separating existing data sets into adults and kids. Three CNN architectures were tested, and the SE-ResNeXt50(32×4d) achieved the highest accuracy at 79.42%. The age-based model outperformed the non-age-based model by 22.24%. The study highlights the impact of age-biased learning data and algorithm types on emotion recognition accuracy, particularly for fear and neutral emotions.

Published

2023-04-15

How to Cite

Hyungjoo Park, Youngha Shin, Kyu Song, Channyeong Yun, & Dongyoung Jang. (2023). Analyzing the Importance of Age-biased Data in Recognizing Emotions from Facial Expressions Using Custom Data Sets and CNN Algorithm. Recent Progress in Science and Technology Vol. 9, 28–45. https://doi.org/10.9734/bpi/rpst/v9/5026B

Issue

Recent Progress in Science and Technology Vol. 9

Section

Chapters