EUVIMEDEUVIMED entdecken
Crossref

Facial Expression Recognition Using Inception-Based AlexNet with Transfer Learning

Xiangmin CHEN, CHAO SHEN, Wenyi LI, Haifeng GUO, Yuling LIANG

IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences · 2026

DOI: 10.1587/transfun.2026eap1071 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, 2026. Verfasst von Xiangmin CHEN und 4 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Facial Expression Recognition plays a crucial role in clinical assessment and personalized treatment planning, particularly for patients who are unable to communicate their pain levels effectively. However, existing expression recognition methods often suffer from insufficient feature extraction capability and limited recognition accuracy due to the complexity and subtlety of pain-related facial expressions. To address these challenges, this paper proposes an Inception-enhanced AlexNet-based deep learning framework for facial expression recognition. By integrating Inception modules into the conventional AlexNet architecture, the proposed network is able to capture multi-scale and more abstract feature representations, thereby improving its feature extraction capability. In addition, transfer learning is employed to initialize the network with pretrained parameters, which significantly accelerates convergence and reduces computational cost during training. Furthermore, fine-tuning of high-level network layers enables the model to better focus on discriminative characteristics specific to facial expressions. Experimental evaluations conducted on three publicly available pain datasets-COPE, UNBC-McMaster, and the BioVid Heat Pain Database-demonstrate that the proposed method outperforms existing approaches in terms of feature extraction effectiveness and classification accuracy. These results confirm the effectiveness of the proposed framework and its potential applicability in practical pain assessment systems.

Zitieren

APA

CHEN, X., SHEN, C., LI, W., GUO, H., & LIANG, Y. (2026). Facial Expression Recognition Using Inception-Based AlexNet with Transfer Learning. IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences. https://doi.org/10.1587/transfun.2026eap1071

Vancouver

CHEN X, SHEN C, LI W, GUO H, LIANG Y. Facial Expression Recognition Using Inception-Based AlexNet with Transfer Learning. IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences. 2026. doi:10.1587/transfun.2026eap1071

BibTeX
@article{CHEN2026,
  author = {CHEN, Xiangmin and SHEN, CHAO and LI, Wenyi and GUO, Haifeng and LIANG, Yuling},
  title = {Facial Expression Recognition Using Inception-Based AlexNet with Transfer Learning},
  journal = {IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences},
  year = {2026},
  doi = {10.1587/transfun.2026eap1071}
}
Kostenlos starten →