EEG Tabanlı Tanıdık ve Tanıdık Olmayan Yüz Tanımada Kanal Seçim Yöntemlerinin Karşılaştırmalı Analizi

Öz

Bu çalışma, tanıdık ve yabancı yüzlerin EEG sinyallerini sınıflandırmak için kanal seçimi ve makine öğrenimine dayalı bir yaklaşım önermektedir. İlk aşamada, Ayrık Dalgacık Dönüşümü kullanılarak EEG sinyallerinden zaman-frekans tabanlı özellikler elde edilmiştir. Daha sonra, Minimum Yedeklilik Maksimum Alaka Düzeyi, Fisher Skoru ve Karşılıklı Bilgi yaklaşımları kullanılarak ayrı EEG kanal alt kümeleri belirlenmiştir. Seçilen kanal özellikleri, Destek Vektör Makinesi sınıflandırıcı kullanılarak sınıflandırılmış ve doğruluk, duyarlılık ve özgüllük performans metrikleri kullanılarak karşılaştırma değerlendirmeleri gerçeklenmiştir. Sonuçlar, sınıflandırma performansının kanal sayısının artmasıyla iyileştiğini göstermektedir. Tüm kanalların kullanılması yaklaşık %91 doğruluk sağlamıştır. Ayrıca, kanal puanlama sonuçları, merkezi, parietal ve posterior bölgelerde bulunan EEG kanallarının tanıdık ve yabancı yüzleri ayırt etmede daha yüksek ayrım gücüne sahip olduğunu göstermektedir. Çalışma sonuçları, düşük sayıda kanalla başarılı tanıdık/tanımadık yüz sınıflandırmasının mümkün olduğunu göstermekte ve taşınabilir EEG tabanlı biyometrik sistemlerin geliştirilmesine katkıda bulunmaktadır.

Anahtar Kelimeler

• EEG
• Tanıdık/tanımadık yüz sınıflandırma
• kanal seçimi
• SVM

Comparative Analysis of Channel Selection Methods for EEG-Based Familiar and Unfamiliar Face Recognition

Öz

This study proposes a channel selection and machine learning-based approach for classifying EEG signals of familiar and unfamiliar faces. In the first stage, time-frequency based features were obtained from EEG signals using Discrete Wavelet Transform. Then, utilizing minimum redundancy maximum relevance, Fisher Score and Mutual Information approaches, separate EEG channel subsets were identified. The chosen channel features were classified using a Support Vector Machine classifier and comparison evaluations were performed utilizing accuracy, sensitivity and specificity performance metrics. The results showed that the classification performance was enhanced with increasing of channel number. Using all channels yielded around 91% accuracy. Furthermore, channel scoring results indicated that EEG channels located in the central, parietal, and posterior regions had higher discrimination in distinguishing between familiar and unfamiliar faces. The study results demonstrate that successful familiar/unfamiliar face classification is possible with a low number of channels and contribute to the development of portable EEG-based biometric systems.

Anahtar Kelimeler

• EEG
• Familiar/Unfamiliar face classification
• channel selection
• SVM

Kaynakça

1. F. K. Bardak and F. Temurtaş, “A Review of EEG-Based Face Recognition: Methodologies, Feature Extraction Techniques, and Classification Methods”, AINTELIA Science Notes Journal, vol. 3, 2024.
2. F. K. Bardak Özkul and F. Temurtaş, “EEG-Based Familiar and Unfamiliar Face Analysis: Classifıcation Using RMS Features with Artificial Neural Networks and Random Forests”, Uludağ Univ. J. Fac. Eng., vol. 30, no. 3, pp. 765-778, 2025.
3. C. Chen, L. Fan, Y. Gao, S. Qiu, W. Wei, and H. He, “EEG-FRM: a neural network based familiar and unfamiliar face EEG recognition method”, Cogn. Neurodyn., vol. 18, no. 2, pp. 357-370, 2024.
4. F. K. Bardak and F. Temurtaş, “Regional Brain Analysis and Machine Learning Techniques for Classifying Familiar and Unfamiliar Faces Using EEG”, Arab. J. Sci. Eng., vol. 50, no. 19, pp. 15551-15576, 2025.
5. G. Liu, Y. Wen, J. H. Hsiao, D. Zhang, L. Tian, and W. Zhou, “EEG-Based Familiar and Unfamiliar Face Classification Using Filter-Bank Differential Entropy Features”, IEEE Trans. Hum.-Mach. Syst., vol. 54, no. 1, pp. 44-55, 2024.
6. Abdullah, I. Faye, and M. R. Islam, “EEG Channel Selection Techniques in Motor Imagery Applications: A Review and New Perspectives”, Bioengineering, vol. 9, no. 12, p. 726, 2022.
7. H. Peng, F. Long, and C. Ding, “Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy”, IEEE Trans. Pattern Anal. Mach. Intell., vol. 27, no. 8, pp. 1226-1238, 2005.
8. M. Qatmh, T. Bonny, N. Nasir, M. Al-Shabi, and A. Al-Shammaa, “Detection of epileptic seizure using discrete wavelet transform on gamma band and artificial neural network,” in 2021 14th International Conference on Developments in eSystems Engineering (DeSE), 2021, pp. 401–406.

9. E. Guttmann-Flury, X. Sheng, and X. Zhu, “Channel selection from source localization: A review of four EEG-based brain–computer interfaces paradigms”, Behav. Res. Methods, vol. 55, no. 4, pp. 1980-2003, 2022.
10. A. K. Dwivedi, O. P. Verma, and S. Taran, “EEG-Based Emotion Recognition Using Optimum Number of Channels”, in 2024 IEEE International Conference on Smart Power Control and Renewable Energy (ICSPCRE), Rourkela, India: IEEE, 2024, pp. 1-6.
11. A. Tiwari and A. Chaturvedi, “Automatic EEG channel selection for multiclass brain-computer interface classification using multiobjective improved firefly algorithm”, Multimed. Tools Appl., vol. 82, no. 4, pp. 5405-5433, 2023.
12. Z.-M. Wang, S.-Y. Hu, and H. Song, “Channel Selection Method for EEG Emotion Recognition Using Normalized Mutual Information”, IEEE Access, vol. 7, pp. 143303-143311, 2019.
13. A. Goshvarpour and A. Goshvarpour, “Lemniscate of Bernoulli’s map quantifiers: innovative measures for EEG emotion recognition”, Cogn. Neurodyn., vol. 18, no. 3, pp. 1061-1077, 2024.
14. X. Yan, A. Volfart, and B. Rossion, “A neural marker of the human face identity familiarity effect”, Sci. Rep., vol. 13, no. 1, p. 16294, 2023.
15. K. Gorur, M. Bozkurt, M. Bascil, and F. Temurtas, “GKP Signal Processing Using Deep CNN and SVM for Tongue-Machine Interface”, Trait. Signal, vol. 36, no. 4, pp. 319-329, 2019.
16. L. Hui, Z. Pei, S. Quan, X. Ke, and S. Zhe, “Cognitive Workload Detection of Air Traffic Controllers Based on mRMR and Fewer EEG Channels”, Brain Sci., vol. 14, no. 8, p. 811, 2024.
17. P. A. Mundra and J. C. Rajapakse, “SVM-RFE With MRMR Filter for Gene Selection”, IEEE Trans. NanoBioscience, vol. 9, no. 1, pp. 31-37, 2010.
18. D. Kaya, “The mRMR-CNN based influential support decision system approach to classify EEG signals.” Measurement, vol. 156, p.107602, 2020.
19. Y. Luo et al., “An EEG channel selection method for motor imagery based on Fisher score and local optimization”, J. Neural Eng., vol. 21, no. 3, p. 036030, 2024.
20. L. Piho and T. Tjahjadi, “A Mutual Information Based Adaptive Windowing of Informative EEG for Emotion Recognition”, IEEE Trans. Affect. Comput., vol. 11, no. 4, pp. 722-735, 2020.