Research on Facial Expression Recognition Based on Improved VGG19

Wei Liang

doi:10.62051/tezx1j29

Authors

Wei Liang

DOI:

https://doi.org/10.62051/tezx1j29

Keywords:

Facial Expression Recognition; Convolutional Neural Network; Depth wise Separable Convolution; Residual Connection; Deep Learning.

Abstract

Facial expression recognition is of great significance in the fields of computer vision and human-computer interaction. Given the problems of over-fitting, high computational complexity, and vanishing gradient in VGG19, this paper proposes an improved VGG19 network model. By introducing depth wise separable convolution, the number of parameters of the model is significantly reduced and the computational efficiency is improved. This paper applies Batch Normalization and Dropout technology to effectively alleviate the over-fitting problem and improve the model's ability for generalization. In addition, this paper introduces residual connections into the model, which solves the vanishing gradient problem and improves the training speed and model stability. This paper uses the FER2013 dataset to conduct comparative experiments on the improved VGG19 network model, the original model, and other network models. The results show that the parameter amount of the improved VGG19 decreased by approximately 45.26%, and the accuracy increased by 4.52%. In addition, the improved VGG19 model is also better than other network models implemented in this paper and has better facial expression recognition results.

Downloads

Download data is not yet available.

References

[1] Binyu Z. Analysis of facial expression recognition technology based on deep learning. Electronic Technology, 2024, 53 (03): 68 - 71.

[2] Jiarong Q, Changwei Z. Research and development of behavior recognition technology. Intelligent Computers and Applications, 2017, 7 (04): 24 - 26+30.

[3] Ning Z. Student classroom emotion analysis based on face recognition. Xi'an University of Technology, 2022.

[4] Bo M, Ke M, Yuehui J, et al. DeepHome: A smart home management and control model based on deep learning. Journal of Computer Science, 2018, 41 (12): 2689 - 2701.

[5] Mengyang X. Improved expression recognition method based on attention mechanism. Modern Information Technology, 2024, 8 (08): 102 - 105+110.

[6] Zhuangzhuang G, Xuebin X, Longbin L, et al. Research on facial expression recognition method based on DenseNet. Computers and Digital Engineering, 2023, 51 (10): 2425 - 2430.

[7] Ye Z, Cihui Y, Jiemei Z, et al. Research on facial expression recognition algorithm based on SR-VGG19. Computers and Digital Engineering, 2021, 49 (09): 1889 - 1894+1898.

[8] Nan Z. Research on facial expression recognition algorithm based on VGG19. Chang'an University, 2023.

[9] Kowsar I, Zaman M S, Sakib M F R. Facial recognition expression: convolutional attentional masking network and ensemble approach. Brac University, 2021.

[10] Yan W, Zhenyu W. Hyperspectral image classification based on improved SE-Net and depthwise separable residuals. Journal of Lanzhou University of Technology, 2024, 50 (02): 87 - 95.

[11] Khaireddin Y, Chen Z. Facial emotion recognition: State of the art performance on FER2013. arXiv preprint arXiv:2105.03588, 2021.

[12] Gang L, Taibing C, Zhibo Y, et al. MBRNet: Multi-branch handwritten character recognition network integrating residual connections. Computer Engineering and Applications, 2024, 1 - 12.