Design and implementation of facial expression recognition using CNN
International Research Journal of Modernization in Engineering Technology & Science · 20.05.2026
DOI: 10.56726/irjmets98550 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Research Journal of Modernization in Engineering Technology & Science, online erschienen am 20.05.2026.
Abstract (Originalsprache)
The study details the design and implementation of a facial expressions recognition (FER) system on the basis of deep learning, which consists of the convolutional neural networks (CNNs).The paper will categorize seven fundamental human emotions of anger, disgust, fear, happiness, sadness, surprise, and neutral by using the FER-2013 data.The four neural network models: a control CNN, a deeper CNN, a transfer learning model based on the VGG19, and a fully connected Artificial Neural Network (ANN) were obtained and tested.Hyperparameter tuning, data preprocessing, and augmentation were done to enhance the performance of models.Findings indicate that convolutional models, especially the Deeper CNN are superior to fully connected models and they can compete with other state-of-the-art studies in terms of accuracy.These results suggest the usefulness of hierarchical feature extraction in recognizing facial emotions and offer details to implement it in human/computer interaction, driver-assistance, mental-health monitoring, and adaptive-learning setting.
Zitieren
APA
(2026). Design and implementation of facial expression recognition using CNN. International Research Journal of Modernization in Engineering Technology & Science. https://doi.org/10.56726/irjmets98550
Vancouver
Design and implementation of facial expression recognition using CNN. International Research Journal of Modernization in Engineering Technology & Science. 2026. doi:10.56726/irjmets98550
BibTeX
@article{euvimed2026,
title = {Design and implementation of facial expression recognition using CNN},
journal = {International Research Journal of Modernization in Engineering Technology & Science},
year = {2026},
doi = {10.56726/irjmets98550}
}
EUVIMED