EUVIMEDEUVIMED entdecken
Crossref

Comparative Performance Analysis of Deep Learning Models for Accurate and robust Face Recognition

Amel Alserry, Mokhtar Ghilan, Najran Al Dawla, Khaled Alalayah

Journal of Eye-Innovation in Machine Learning · 12.08.2026 · 1 (20261) · 54

DOI: 10.68990/jieml.202613 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Eye-Innovation in Machine Learning, online erschienen am 12.08.2026. Verfasst von Amel Alserry und 3 weiteren Autor:innen. Die Arbeit nennt 98 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Face recognition has emerged as a vital biometric technology essential across numerous domains, including electronic security, digital forensics, healthcare, and automated access control systems. Classifying and verifying human identities using facial imagery represents a critical task in personal authentication, yet maintaining high accuracy remains challenging due to real-world variations such as lighting fluctuations, extreme pose angles, and partial occlusions. Face recognition using digital image processing and computer vision has become an essential research area for non-invasive, cost-effective biometric identification. The application of diverse deep learning architectures, combined with feature extraction techniques, has been the primary focus of recent research aimed at enhancing system reliability. In this study, deep learning models alongside traditional feature descriptors— including Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), DualCross Patterns (DCP), and Trunk-Branch Ensemble CNNs—were integrated and evaluated to optimize feature representation and recognition robustness. The models were trained and benchmarked across diverse facial image conditions. The experimental findings demonstrated that multimodal deep learning representations and ensemble convolutional neural networks significantly improved classification accuracy and robustness to noise compared with traditional handcrafted algorithms in unconstrained operational environments. The proposed methodology encompasses facial image pre-processing, landmark identification, robust feature extraction using deep neural networks, and classification based on distinct facial texture and geometric patterns.

Zitieren

APA

Alserry, A., Ghilan, M., Dawla, N. A., & Alalayah, K. (2026). Comparative Performance Analysis of Deep Learning Models for Accurate and robust Face Recognition. Journal of Eye-Innovation in Machine Learning, 1(20261), 54. https://doi.org/10.68990/jieml.202613

Vancouver

Alserry A, Ghilan M, Dawla NA, Alalayah K. Comparative Performance Analysis of Deep Learning Models for Accurate and robust Face Recognition. Journal of Eye-Innovation in Machine Learning. 2026;1(20261):54. doi:10.68990/jieml.202613

BibTeX
@article{Alserry2026,
  author = {Alserry, Amel and Ghilan, Mokhtar and Dawla, Najran Al and Alalayah, Khaled},
  title = {Comparative Performance Analysis of Deep Learning Models for Accurate and robust Face Recognition},
  journal = {Journal of Eye-Innovation in Machine Learning},
  year = {2026},
  volume = {1},
  number = {20261},
  pages = {54},
  doi = {10.68990/jieml.202613}
}
Kostenlos starten →