Optimization in Facial Expression Recognition Based on CNN Combined with SE Modules

DOI

10.2991/978-94-6463-540-9_99How to use a DOI?

Keywords

Facial expression recognition; convolutional neural networks; attention mechanisms

Abstract

Facial expression recognition has emerged as a pivotal aspect of human-computer interaction and psychological research, drawing extensive attention in computer vision. The essay aims to improve the facial expression recognition performance of Convolutional Neural Networks (CNN) under different imaging conditions by combining attention mechanisms. In terms of data preparation, the FER-2013 dataset from Kaggle was used, which includes grayscale facial images with 48 x 48 pixels. By data augmentation and normalization, the diversity of the data is increased, and the robustness of the model is improved through random horizontal flipping, brightness and contrast adjustment, and the introduction of Gaussian noise. In terms of model architecture, a network structure similar to VGG is adopted, and a Squeeze and Excitation (SE) module is introduced after each convolutional layer, dynamically adjusting the importance of each channel through global average pooling and fully connected layers. The experimental results indicate that incorporating the attention mechanism reduces the model’s loss across the training, validation, and test sets, while significantly enhancing its accuracy. These results demonstrate the effectiveness of attention mechanisms in facial expression recognition tasks. Overall, this study significantly improved the performance and robustness of CNN in facial expression recognition tasks by introducing attention mechanisms, demonstrating its superiority under complex imaging conditions.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Xuanyu Zhang
PY  - 2024
DA  - 2024/10/16
TI  - Optimization in Facial Expression Recognition Based on CNN Combined with SE Modules
BT  - Proceedings of the 2024 2nd International Conference on Image, Algorithms and Artificial Intelligence (ICIAAI 2024)
PB  - Atlantis Press
SP  - 993
EP  - 1002
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-540-9_99
DO  - 10.2991/978-94-6463-540-9_99
ID  - Zhang2024
ER  -