Journal of Science and Innovative Development ISSN 2181-4317

FACIAL EMOTION DETECTION USING PRE-TRAINED ADVANCED CONVOLUTIONAL NEURAL NETWORKS AND THE IMPROVED FER- 2013 DATASET

Kurbanov Abduraxmon Alishboyevich March 8, 2025 DOI: https://doi.org/10.36522/ILM-FAN/8-2-2025-4b0d5

Abstract

<p>&nbsp;Emotion recognition from human facial images is one of the important&nbsp;<br /> directions for human-computer interaction, security systems, mental health&nbsp;<br /> monitoring, and intelligent systems. Especially in the development of humanoid&nbsp;<br /> robots in the field of human-computer interaction, one of the important features&nbsp;<br /> of a humanoid robot is that the robot can communicate with humans while sensing&nbsp;<br /> their emotions. The development of emotion recognition systems is a significant&nbsp;<br /> challenge in computer vision and deep learning. In this study, we suggest a good&nbsp;<br /> way to recognize emotions from people&#39;s facial images by using well-known&nbsp;<br /> convolutional neural networks that have been trained on a better version of&nbsp;<br /> the FER-2013 dataset. In this study, the effectiveness of implementing the task&nbsp;<br /> of emotional state recognition using advanced convolutional neural networks&nbsp;<br /> is studied. In particular, the results of the popular pre-trained architectures&nbsp;<br /> ResNet-50, VGGNet-16, DenseNet-121, and EfficientNet-B0 on facial expression&nbsp;<br /> recognition are analyzed. The study used an improved and augmented version&nbsp;<br /> of the FER-2013 dataset. During data processing, imbalances between facial&nbsp;<br /> expressions, low-quality images, and incorrectly labeled images were detected&nbsp;<br /> and corrected. In addition, data augmentation techniques were used to reduce the&nbsp;<br /> problem of overfitting. The model results were evaluated based on criteria such as&nbsp;<br /> accuracy and loss function.</p>

Cite this article
Kurbanov Abduraxmon Alishboyevich (2025). FACIAL EMOTION DETECTION USING PRE-TRAINED ADVANCED CONVOLUTIONAL NEURAL NETWORKS AND THE IMPROVED FER- 2013 DATASET. Journal of Science and Innovative Development. https://doi.org/https://doi.org/10.36522/ILM-FAN/8-2-2025-4b0d5