Comparison and Analysis of Facial Emotion Detection Using Various Deep Learning Neural Networks

Notice

This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.

Year : 2025 | Volume : 12 | Issue : 02 | Page : –
    By

    V. Anusuya,

  • S. Sakkaravarthi,

  • M. Rethinakumari,

  • Sridhar S.,

  1. Associate Professor, Department of Information Technology, Ramco institute of Technology, Rajapalayam, Tamil Nadu, India
  2. Associate Professor, Department of Information Technology, Ramco institute of Technology, Rajapalayam, Tamil Nadu, India
  3. Associate Professor, Department of Information Technology, Ramco institute of Technology, Rajapalayam, Tamil Nadu, India
  4. Student, Third year, Department of Information Technology, Ramco institute of Technology, Rajapalayam, Tamil Nadu, India

Abstract

document.addEventListener(‘DOMContentLoaded’,function(){frmFrontForm.scrollToID(‘frm_container_abs_186607’);});Edit Abstract & Keyword

Facial emotion recognition employs Convolutional Neural Networks (CNNs), Residual Networks (ResNet), Long Short-Term Memory (LSTM) networks, and Deep Neural Networks (DNNs) to automatically identify various emotions, including disgust, anger, fear, happiness, sadness, surprise, and neutrality. This study utilizes transfer learning along with data preprocessing techniques such as rotation, flipping, brightness adjustment, and enhancement methods. Traditional machine learning models achieve an accuracy range of 45% to 50%. In contrast, our proposed CNN and DNN models show improved accuracy, reaching 65% and 62%, respectively. Additionally, we introduce a hybrid model combining ResNet and LSTM architecture, which achieves an accuracy of 72%. Following this, we conduct a comparative analysis of the accuracy and loss for each model. Our findings indicate that while CNNs initially demonstrated higher accuracy than the hybrid ResNet-LSTM model, the hybrid model ultimately surpassed all others in total classification accuracy. The performance metrics used in this analysis include recall and precision.

Keywords: Facial emotion recognition (FER), neural networks, convolutional neural networks (CNNs), residual networks (ResNet), long short-term memory (LSTM)

[This article belongs to Journal of Image Processing & Pattern Recognition Progress ]

How to cite this article:
V. Anusuya, S. Sakkaravarthi, M. Rethinakumari, Sridhar S.. Comparison and Analysis of Facial Emotion Detection Using Various Deep Learning Neural Networks. Journal of Image Processing & Pattern Recognition Progress. 2025; 12(02):-.
How to cite this URL:
V. Anusuya, S. Sakkaravarthi, M. Rethinakumari, Sridhar S.. Comparison and Analysis of Facial Emotion Detection Using Various Deep Learning Neural Networks. Journal of Image Processing & Pattern Recognition Progress. 2025; 12(02):-. Available from: https://journals.stmjournals.com/joipprp/article=2025/view=0



document.addEventListener(‘DOMContentLoaded’,function(){frmFrontForm.scrollToID(‘frm_container_ref_186607’);});Edit

References

  1. Yaddaden Y. An efficient facial expression recognition system with appearance-based fused descriptors. Intelligent Systems with Applications. 2023 Feb 1;17:200166.
  2. Gupta S, Kumar P, Tekchandani RK. Facial emotion recognition based real-time learner engagement detection system in online learning context using deep learning models. Multimedia Tools and Applications. 2023 Mar;82(8):11365–94.
  3. Dias W, Andalo F, Padilha R, Bertocco G, Almeida W, Costa P, Rocha A. Cross-dataset emotion recognition from facial expressions through convolutional neural networks. Journal of Visual Communication and Image Representation. 2022 Jan 1;82:103395.
  4. Sajjad M, Ullah FU, Ullah M, Christodoulou G, Cheikh FA, Hijji M, Muhammad K, Rodrigues JJ. A comprehensive survey on deep facial expression recognition: challenges, applications, and future guidelines. Alexandria Engineering Journal. 2023 Apr 1;68:817–40.
  5. Meena G, Mohbey KK, Indian A. Categorizing sentiment polarities in social networks data using convolutional neural network. SN Computer Science. 2022 Mar;3(2):116.
  6. Castellano G, De Carolis B, Macchiarulo N. Automatic facial emotion recognition at the COVID- 19 pandemic time. Multimedia Tools and Applications. 2023 Apr;82(9):12751–69.
  7. Elsheikh RA, Mohamed MA, Abou-Taleb AM, Ata MM. Improved facial emotion recognition model based on a novel deep convolutional structure. Scientific Reports. 2024 Nov 23;14(1):29050.
  8. Kim JC, Kim MH, Suh HE, Naseem MT, Lee CS. Hybrid approach for facial expression recognition using convolutional neural networks and SVM. Applied Sciences. 2022 May 28;12(11):5493.
  9. Li B, Lima D. Facial expression recognition via ResNet-50. International Journal of Cognitive Computing in Engineering. 2021 Jun 1;2:57–64.
  10. Shehu HA, Sharif MH, Uyaver S. Facial expression recognition using deep learning. InAIP Conference Proceedings 2021 Mar 2 (Vol. 2334, No. 1). AIP Publishing.
Regular Issue Subscription Original Research
Volume 12
Issue 02
Received 06/03/2025
Accepted 06/04/2025
Published 28/04/2025
Publication Time 53 Days
244

[first_name] [last_name]

My IP

PlumX Metrics