Impact of Image Quality Enhancement Using Homomorphic Filtering on the Performance of Deep Learning-Based Facial Emotion Recognition Systems

Authors

  • Al Bahri Department of Electrical and Computer Engineering, Universitas Syiah Kuala
  • Maulisa Oktiana Department of Electrical and Computer Engineering, Universitas Syiah Kuala
  • Maya Fitria Department of Electrical and Computer Engineering, Universitas Syiah Kuala https://orcid.org/0000-0002-6593-212X
  • Zulfikar Zulfikar Department of Electrical and Computer Engineering, Universitas Syiah Kuala

DOI:

https://doi.org/10.26555/jiteki.v11i2.30409

Abstract

Facial emotion recognition technology is crucial in understanding human expressions from images or videos by analyzing distinct facial features.  A common challenge in this technology is accurately detecting a person's facial expression, which can be hindered by unclear facial lines, often due to poaor lighting conditions. To address these challenges, it is essential to improve image quality. This study investigates how enhancing image quality through homomorphic filtering and sharpening techniques can boost the accuracy and performance of deep learning-based facial emotion recognition systems. Improved image quality allows the classification model to focus on relevant expression features better. The testing used three different architectures: MobileNet, InceptionV3, and DenseNet121. Experimental results indicated that image enhancement positively impacts the accuracy of the facial emotion recognition system. Specifically, the average accuracy increased by 1-2% for the MobileNet architecture, by 5-7% for InceptionV3, and by 1-3% for DenseNet121.

Published

2025-04-28

How to Cite

[1]
A. Bahri, M. Oktiana, M. Fitria, and Z. Zulfikar, “Impact of Image Quality Enhancement Using Homomorphic Filtering on the Performance of Deep Learning-Based Facial Emotion Recognition Systems”, J. Ilm. Tek. Elektro Komput. Dan Inform, vol. 11, no. 2, Apr. 2025.

Issue

Vol. 11 No. 2 (2025): June

Section

Articles

License

Copyright (c) 2025 Al Bahri, Maulisa Oktiana, Maya Fitria, Zulfikar Zulfikar

Creative Commons License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Authors who publish with JITEKI agree to the following terms:

  1. Authors retain copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (CC BY-SA 4.0) that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
  2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
  3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.

Lisensi Creative Commons

This work is licensed under a Creative Commons Attribution 4.0 International License