Journals / Sinop Üniversitesi Fen Bilimleri Dergisi / 2025 / Cilt: 10 Sayı: 2

Artificial Intelligence-Based Real-Time Driver Drowsiness Detection: A Hybrid Model Approach

Yapay Zekâ Tabanlı Gerçek Zamanlı Sürücü Uyuşukluğu Tespiti: Hibrit Model Yaklaşımı

Gulay Cicek(BEYKENT ÜNİVERSİTESİ)0000-0002-6607-1181
Pages
621–651
Publication Date
2025-12-24
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

In conclusion, the proposed deep learning-based model demonstrates significant potential in transforming traffic safety by effectively detecting driver fatigue.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:This study presents the development of a deep learning-based model designed to detect fatigue and drowsiness of drivers in real time.
📊
Key Findings & Data:The MRL Eye Dataset, containing approximately 85.000 images, is used to classify eye states, while the Yawn Dataset, with over 5.000 images, supports the detection of yawning movements.
💡
Conclusion & Impact:In conclusion, the proposed deep learning-based model demonstrates significant potential in transforming traffic safety by effectively detecting driver fatigue.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Shallow water denklemi · mertebe indirgenmiş modelleme · düzgün dik ayrışma

Abstract

This study presents the development of a deep learning-based model designed to detect fatigue and drowsiness of drivers in real time. Utilizing open-access datasets such as the MRL Eye Dataset and the Yawn Dataset, the model performs an in-depth analysis of eye open/closed states and yawning behaviors of drivers. The MRL Eye Dataset, containing approximately 85.000 images, is used to classify eye states, while the Yawn Dataset, with over 5.000 images, supports the detection of yawning movements. To enhance model accuracy, these datasets are balanced and optimized through various preprocessing techniques. The model is trained using Convolutional Neural Networks (CNNs) and further improved through transfer learning methods, significantly increasing classification performance. The proposed model achieves impressive metrics with an accuracy of 98%, a precision of 97.5%, and a specificity of 98.2%, indicating high performance. By focusing on blink and yawn detection, this study offers a novel approach compared to existing literature and provides a more reliable and effective solution to detect driver fatigue. Moreover, the use of synthetic data allows the model to be trained with broader and more diverse datasets, overcoming the limitations of real data collection. In future work, incorporating additional biometric indicators such as head movements and facial expressions could further improve the model’s accuracy and enable a more comprehensive assessment of driver alertness. Additionally, the generalizability of the model can be enhanced by including data from different cultural and geographical groups, thereby extending its applicability to a wider range of users. In conclusion, the proposed deep learning-based model demonstrates significant potential in transforming traffic safety by effectively detecting driver fatigue. With the advancement of technologies such as autonomous vehicles and intelligent transportation systems, this model could be integrated into driver assistance systems to help prevent accidents and enhance guest safety. It represents a revolutionary step toward preventing traffic accidents and ensuring driver well-being.

Özet

Sürücülerin yorgunluk ve uyuşukluk durumlarını gerçek zamanlı olarak tespit etmeyi amaçlayan derin öğrenme tabanlı bir model geliştirmektedir. MRL Eye Dataset ve Yawn Dataset gibi açık erişimli veri setlerinden faydalanarak, sürücülerin göz açık/kapalı durumları ve esneme hareketleri derinlemesine analiz edilmektedir. MRL Eye Dataset, yaklaşık 85.000 görüntü ile sürücünün göz durumlarını sınıflandırırken, Yawn Dataset ise 5.000’den fazla görüntü ile esneme hareketlerini tespit etmek için kullanılmaktadır. Bu veri setleri, modelin doğruluğunu artırmak amacıyla dengeli şekilde düzenlenmekte ve çeşitli ön işleme teknikleri ile iyileştirilmektedir. Model, CNN ile eğitilmekte ve transfer öğrenme teknikleriyle güçlendirilmekte, bu sayede modelin sınıflandırma başarısı önemli ölçüde artırılmaktadır. Modelin elde ettiği doğruluk oranı %98, hassasiyet oranı %97.5 ve özgüllük oranı ise %98.2 gibi yüksek metriklerle başarılı sonuçlar elde edilmektedir. Bu çalışma, literatürde ilk kez hem göz kırpma hem de esneme hareketlerini hibrit CNN tabanlı bir modelde birlikte ele almaktadır. Bu özgün yaklaşım, sürücü yorgunluğunu tespit etmede yalnızca tek parametreye odaklanan çalışmalara kıyasla daha kapsamlı ve güvenilir sonuçlar üretmektedir. Ayrıca, sentetik verilerin kullanılması, gerçek veri toplama zorluklarını aşarak daha geniş ve çeşitlendirilmiş veri setleriyle modelin eğitilmesine olanak tanımaktadır. Gelecekte, baş hareketleri, yüz ifadeleri ve diğer biyometrik verilerin sisteme entegre edilmesi ile modelin doğruluğu daha da artırılabilir ve sürücülerin dikkat seviyelerini daha kapsamlı bir şekilde değerlendirebilir. Ayrıca, farklı kültürel ve coğrafi gruplardan elde edilen verilerle modelin genellenebilirliği sağlanarak, daha geniş bir kullanıcı kitlesine hitap edilmesi mümkün hale gelebilmektedir. Sonuç olarak, bu çalışma, sürücü yorgunluğunu tespit etmeye yönelik geliştirilen derin öğrenme tabanlı modelin, trafik güvenliğini derinlemesine dönüştürebilecek büyük bir potansiyele sahip olduğunu ve yol güvenliğini önemli ölçüde artırabileceğini ortaya koymaktadır. Özellikle otonom araçlar ve akıllı ulaşım sistemleri gibi teknolojilerin gelişimiyle paralel olarak, bu model sürücü destek sistemlerine entegre edilerek, kazaların önlenmesine katkı sağlayabilir ve sürücülerin güvenliğini artırabilir. Bu model, trafik kazalarını önlemek ve sürücülerinin güvenliğini sağlamak adına devrim niteliğinde bir adım olarak değerlendirilebilmektedir.

🔖

Cite This Article

Terzi, F., Süer, Ö., Cicek, G. (2025). Artificial Intelligence-Based Real-Time Driver Drowsiness Detection: A Hybrid Model Approach. Sinop Üniversitesi Fen Bilimleri Dergisi, Cilt: 10 Sayı: 2, 621–651. https://doi.org/10.33484/sinopfbd.1701924

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (35)

  1. Majeed, F., Shafique, U., Safran, M., Alfarhood, S., & Ashraf, I. (2023). Detection of drowsiness among drivers using novel deep convolutional neural network model. Sensors, 23(21), 8741. https://doi.org/10.3390/s23218741
  2. Jahan, I., Uddin, K. M. A., Murad, S. A., Miah, M. S. U., Khan, T. Z., Masud, M., Aljahdali, S., & Bairagi, A. K. (2023). 4d: A real-time driver drowsiness detector using deep learning. Electronics 12(1), 235. 10.3390/electronics12010235
  3. Florez, R., Palomino-Quispe, F., Coaquira-Castillo, R. J., Herrera-Levano, J. C., Paixão, T., & Alvarez, A. B. (2023). A cnn-based approach for driver drowsiness detection by real-time eye state identification. Applied Sciences, 13(13), 7849. https://doi.org/10.3390/app13137849
  4. Beles, H., Vesselenyi, T., Rus, A., Mitran, T., Scurt, F. B., & Tolea, B. A. (2024). Driver drowsiness multi-method detection for vehicles with autonomous driving functions. Sensors, 24(5), 1541. https://doi.org/10.3390/s24051541
  5. Amidei, A., Spinsante, S., Iadarola, G., Benatti, S., Tramarin, F., Pavan, P., & Rovati, L. (2023). Driver drowsiness detection: A machine learning approach on skin conductance. Sensors, 23(8), 4004. https://doi.org/10.3390/s23084004
  6. Wu, F., Fu, R., Ma, Y., Wang, C., & Zhang, Z. (2020). Relationship between speed perception and eye movement—a case study of crash-involved and crash-not-involved drivers in china. Plos one, 15(3), e0229650.
  7. Ebrahim Shaik, M. (2023). A systematic review on detection and prediction of driver drowsiness. Transportation Research Interdisciplinary Perspectives, 21, 100864. https://doi.org/10.1016/j.trip.2023.100864
  8. Das, S., Pratihar, S., Pradhan, B., Jhaveri, R. H., & Benedetto, F. (2024). Iot-assisted automatic driver drowsiness detection through facial movement analysis using deep learning and a u-net-based architecture. Information, 15(1), 30. 10.3390/info15010030
  9. Albadawi, Y., AlRedhaei, A., & Takruri, M. (2023). Real-time machine learning-based driver drowsiness detection using visual features. Journal of Imaging, 9(5), 91. https://doi.org/10.3390/jimaging9050091
  10. Shahbakhti, M., Beiramvand, M., Nasiri, E., Far, S. M., Chen, W., Solé-Casals, J., Wierzchon, M., Broniec-Wójcik, A., Augustyniak, P., & Marozas, V. (2023). Fusion of eeg and eye blink analysis for detection of driver fatigue. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 31, 2037–2046. https://doi.org/10.1109/TNSRE.2023.3267114
  11. Safarov, F., Akhmedov, F., Abdusalomov, A. B., Nasimov, R., & Cho, Y. I. (2023). Real-time deep learning-based drowsiness detection: Leveraging computer-vision and eye-blink analyses for enhanced road safety. Sensors, 23(14), 6459. https://doi.org/10.3390/s23146459
  12. Nasri, I., Karrouchi, M., Kassmi, K., & Messaoudi, A. (2022), A review of driver drowsiness detection systems: Techniques, advantages and limitations.
  13. Phan, A.-C., Nguyen, N.-H.-Q., Trieu, T.-N., & Phan, T.-C. (2021). An efficient approach for detecting driver drowsiness based on deep learning. Applied Sciences, 11(18), 8441. https://doi.org/10.3390/app11188441
  14. Oh, S.-j., Jung, M.-j., Lim, C., & Shin, S.-c. (2020). Automatic detection of welding defects using faster r-cnn. Applied Sciences, 10(23), 8629. https://doi.org/10.3390/app10238629
  15. Hashemi, M., Mirrashid, A., & Beheshti Shirazi, A. (2020). Driver safety development real time driver drowsiness detection system based on convolutional neural network. arXiv preprint. arXiv:200105137.
  16. Dreissig, M., Baccour, M. H., Schaeck, T., & Kasneci, E. (2020). Driver drowsiness classification based on eye blink and head movement features using the k-nn algorithm. arXiv preprint. arXiv:200913276.
  17. Deng, W., & Wu, R. (2019). Real-time driver-drowsiness detection system using facial features. IEEE Access, 7, 118727–118738. https://doi.org/10.1109/ACCESS.2019.2936663
  18. Magán, E., Sesmero, M. P., Alonso-Weber, J. M., &Sanchis, A. (2022). Driver drowsiness detection by applying deep learning techniques to sequences of images. Applied Sciences, 12(3), 1145. 10.3390/app12031145
  19. Alajlan, N. N., & Ibrahim, D. M. (2023). Ddd tinyml: A tinyml-based driver drowsiness detection model using deep learning. Sensors, 23(12), 5696. https://doi.org/10.3390/s23125696
  20. Voulodimos, A., Doulamis, N., Doulamis, A., & Protopapadakis, E. (2018). Deep learning for computer vision: A brief review. Computational intelligence and neuroscience, 2018(1), 7068349.
  21. Youssouf, N. (2022). Traffic sign classification using cnn and detection using faster-rcnn and yolov4. Heliyon, 8(12), e11792. https://doi.org/10.1016/j.heliyon.2022.e11792
  22. Zhang, T., Liu, X., Shao, M., Sun, Y., & Zhang, Q. (2025). Ship ranging method in lake areas based on binocular vision. Sensors, 25(20), 6477. https://doi.org/10.3390/s25206477
  23. Dewi, C., Chen, R.-C., Chang, C.-W., Wu, S.-H., Jiang, X., & Yu, H. (2022). Eye aspect ratio for real-time drowsiness detection to improve driver safety. Electronics, 11(19), 3183. https://doi.org/10.3390/electronics11193183
  24. Florez, R., Palomino-Quispe, F., Alvarez, A. B., Coaquira-Castillo, R. J., & Herrera-Levano, J. C. (2024). A real-time embedded system for driver drowsiness detection based on visual analysis of the eyes and mouth using convolutional neural network and mouth aspect ratio. Sensors, 24(19), 6261. https://doi.org/10.3390/s24196261
  25. Younes, K., Mouhtady, O., Chaouk, H., Obeid, E., Roufayel, R., Moghrabi, A., & Murshid, N. (2021). The application of principal component analysis (pca) for the optimization of the conditions of fabrication of electrospun nanofibrous membrane for desalination and ion removal. Membranes, 11(12), 979. https://doi.org/10.3390/membranes11120979
  26. Vergni, L., & Todisco, F. (2023). A random forest machine learning approach for the identification and quantification of erosive events. Water, 15(12), 2225. https://doi.org/10.3390/w15122225
  27. Wang, X., Wang, Z., Du, W., Ma, X., Ma, J., Chen, Z., Gao, C., & Chen, X. (2024). Predictive value of tyg and tyg-bmi indices for non-alcoholic fatty liver disease in high-altitude regions of china: A cross-sectional study. Journal of Clinical Medicine, 13(23), 7423. https://doi.org/10.3390/jcm13237423
  28. Ben Salem, F., Almousa, M. T., & Derbel, N. (2024). Direct torque control with space vector modulation (dtc-svm) with adaptive fractional-order sliding mode: A path towards improved electric vehicle propulsion. World Electric Vehicle Journal, 15(12), 563. https://doi.org/10.3390/wevj15120563
  29. Casado, U. M., Altuna, F. I., & Miccio, L. A. (2024). Towards sustainable material design: A comparative analysis of latent space representations in ai models. Sustainability, 16(23), 10681. https://doi.org/10.3390/su162310681
  30. Rezk, N. G., Alshathri, S., Sayed, A., & El-Din Hemdan, E. (2024). Ewais: An ensemble learning and explainable ai approach for water quality classification toward iot-enabled systems. Processes, 12(12), 2771. https://doi.org/10.3390/pr12122771
  31. Meléndez, R., Ptaszynski, M., & Masui, F. (2024). Comparative investigation of traditional machine-learning models and transformer models for phishing email detection. Electronics, 13(24), 4877. https://doi.org/10.3390/electronics13244877
  32. Makhmudov, F., Kutlimuratov, A., & Cho, Y.-I. (2024). Hybrid lstm–attention and cnn model for enhanced speech emotion recognition. Applied Sciences, 14(23), 11342. https://doi.org/10.3390/app142311342
  33. Zhang, H., Zhu, C., Jiao, T., Luo, K., Ma, X., & Wang, M. (2024). Analysis of the trends and driving factors of cultivated land utilization efficiency in henan province from 2000 to 2020. Land, 13(12), 2109. https://doi.org/10.3390/land13122109
  34. Huang, J., & Zhang, C. (2024). Daily tourism demand forecasting with the itransformer model. Sustainability, 16(23), 10678. https://doi.org/10.3390/su162310678
  35. Zacarias, H., Marques, J. A. L., Felizardo, V., Pourvahab, M., & Garcia, N. M. (2024). Ecg forecasting system based on long short-term memory. Bioengineering, 11(1), 89. https://doi.org/10.3390/bioengineering11010089