NeuroGuard: Digital Eye Fatigue Monitoring via Hybrid Landmark Analysis
DOI:
https://doi.org/10.63503/acset.124Keywords:
Artificial Intelligence, CNN–LSTM, MediaPipe, Eye Aspect Ratio (EAR), Percentage of Eye Closure (PERCLOS), Real-Time Monitoring, Driver Drowsiness Detection, Digital Eye FatigueAbstract
Fatigue has a pronounced effect on reaction times, sustained attention, decision-making, and visual performance, and is a serious threat to road traffic safety and occupational productivity. Therefore, real-time and reliable eye fatigue detection systems are key components of intelligent transportation and human-computer interaction technologies. Existing vision-based methods either rely solely on deep learning to extract spatial features without interpretable physiological markers or require manually designed geometric features with conventional classifiers, which have limitations.
In this paper, we propose an efficient real-time eye fatigue monitoring system, called NeuroGuard, based on a hybrid Convolutional Neural Network–Long Short-Term Memory (CNN–LSTM) deep learning architecture and MediaPipe-based facial landmark tracking. The system combines spatial feature representations with temporal eye behavior patterns to provide a method for quantifying blink rate and eyelid closure duration using Eye Aspect Ratio (EAR) and Percentage of Eye Closure (PERCLOS) metrics. Experimental evaluation in three real-world scenarios, including professional driving, long-term software development, and night-shift monitoring under low-light conditions, demonstrates that NeuroGuard achieves 94.3% accuracy, 96.2% recall, and 91.8% precision, improving by 12.2%, 11.2%, and 13.4%, respectively, over traditional threshold-based baselines. Our system reduces false positives from 14.5% to 3.2% and achieves a practical inference latency of approximately 13 ms, making it suitable for safety-critical real-time deployment.
References
[1] A. Guleria and R. Kaur, "Digital Eye Strain Detection System Based on SVM," in Proceedings of the 2021 5th International Conference on Trends in Electronics and Informatics (ICOEI), Tirunelveli, India, June 2021, pp. 1–6. doi: 10.1109/ICOEI51242.2021.9453085
[2] Y. Qin, H. Lyu, and K. Zhu, "Driver fatigue detection method based on multi-feature empirical fusion model," Multimedia Tools and Applications, vol. 83, no. 14, pp. 41275–41296, March 2024. doi: 10.1007/s11042-024-20115-z.
[3] Y. Xiang, "Eye fatigue state recognition of Gabor wavelet optimization HMM algorithm," International Journal of Hybrid Information Technology, vol. 9, no. 9, pp. 189–198, 2016. doi: 10.14257/ijhit.2016.9.9.18.
[4] A. Kuwahara, R. Hirakawa, H. Kawano, K. Nakashi, and Y. Nakatoh, "Eye fatigue prediction system using blink detection based on eye image," in Proceedings of the 2021 IEEE International Conference on Consumer Electronics (ICCE), Las Vegas, NV, USA, January 2021, pp. 1–4. doi: 10.1109/ICCE50685.2021.9427681.
[5] R. Ni, "Research on driver fatigue detection method based on convolutional neural networks using eye features," in Proceedings of the IEEE Conference on Artificial Intelligence and Applications in High Performance Computing (AIAHPC), Wuhan, China, 2025, pp. 191–195.
[6] H. A. Madni, A. Raza, R. Sehar, N. Thalji, and L. Abualigah, "Novel transfer learning approach for driver drowsiness detection using eye movement behavior," IEEE Access, vol. 12, pp. 64765–64778, April 2024. doi: 10.1109/ACCESS.2024.3392640.
[7] L. Xie and S. Fan, "A learning-based multi-frame visual feature framework for real-time driver fatigue detection," in Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (System Demonstrations), Albuquerque, NM, USA, April 2025, pp. 61–69. doi: 10.18653/v1/2025.naacl-demo.
[8] Y. Xiang, R. Hu, Y. Xu, C.-Y. Hsu, and C. Du, "Gaussian weighted eye state determination for driving fatigue detection," Mathematics, vol. 11, no. 9, Article 2101, April 2023. doi: 10.3390/math11092101.
[9] G. F. D. Almeida, Real-Time Detection of Driver Fatigue Using Mobile Device Sensors and Artificial Intelligence On-Device, M.S. Thesis, Department of Informatics Engineering, Faculty of Engineering, University of Porto, Porto, Portugal, July 2025.
[10] M. Vasiljevas, R. Damaševičius, and R. Maskeliūnas, "A human-adaptive model for user performance and fatigue evaluation during gaze-tracking tasks," Electronics, vol. 12, no. 5, Article 1130, February 2023. doi: 10.3390/electronics12051130.
[11] M. S. Chae et al., "Eye-Blink and SSVEP-Based Selective Attention Interface for XR User Authentication: An Explainable Neural Decoding and Machine Learning Approach to Reducing Visual Fatigue," IEEE Access, vol. 13, pp. 176998–177018, 2025. doi: 10.1109/ACCESS.2025.3613355.
[12] S. Das, B. Bhaumik, B. Bagchi, S. Roy, and S. Mukherjee, "A comparative study on procedural and predictive systems on drowsiness detection," in Proceedings of the 2023 7th International Conference on Electronics, Materials Engineering & Nano-Technology (IEMENTech), Kolkata, India, April 2023, pp. 1–5. doi: 10.1109/IEMENTech60402.2023.10423474.
[13] S. Selvi, O. G. Elamathy, and S. Swetha, "Eye fatigue detection: A hybrid CNN–LSTM approach for digital screen users," in Innovations in Computer Vision, Communication Systems, and Computational Intelligence, S. L. Tripathi, O. P. Kumar, A. D. Stalin, and T. Ali, Eds. Boca Raton, FL, USA: CRC Press, 2025, pp. 75–81.
[14] H. Sidaq, L. Wang, S. Guizani, H. Haider, A. U. Rehman, and H. Hamam, "VMFD: Virtual Meetings Fatigue Detector Using Eye Polygon Area and Dlib Shape Indicator," Computers, Materials & Continua, vol. 86, no. 3, p. 25, 2026. doi: 10.32604/cmc.2025.071254.
[15] M. El Harti, S. J. Andaloussi, and O. Ouchetto, "A comprehensive review comparing artificial intelligence and clinical diagnostic approaches for dry eye disease," Diagnostics, vol. 15, no. 23, Article 3071, November 2025. doi: 10.3390/diagnostics15233071.
[16] G. Kiss and P. Viktor, "Artificial intelligence-based depression detection," Sensors, vol. 26, no. 2, Article 748, January 2026. doi: 10.3390/s26020748.
[17] C. F. Blanco-Díaz et al., "Effects of the concentration level, eye fatigue and coffee consumption on the performance of a BCI system based on visual ERP-P300," Journal of Neuroscience Methods, vol. 382, Article 109722, October 2022. doi: 10.1016/j.jneumeth.2022.109722.
[18] I. Kayadibi, G. E. Güraksın, U. Ergün, and N. Öz. Süzme, "An eye state recognition system using transfer learning: AlexNet-based deep convolutional neural network," International Journal of Computational Intelligence Systems, vol. 15, Article 49, 2022. doi: 10.1007/s44196-022-00108-2.
[19] C. Wang, M. Lin, L. Shao, and J. Xiang, "Temporal facial features based fatigue detection system for low-cost devices," Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, vol. 239, no. 14, pp. 7272–7286, February 2025. doi: 10.1177/09544070241306160.
[20] Q. Yang, Z. Li, C. Li, and C. Zhu, "Eye state recognition algorithm GAHMM of web-based learning fatigue," Journal of Multimedia, vol. 9, no. 3, pp. 397–405, 2014. doi: 10.4304/jmm.9.3.397-405.
[21] H. Rahman, M. U. Ahmed, S. Barua, P. Funk, and S. Begum, "Vision-based driver's cognitive load classification considering eye movement using machine learning and deep learning," Sensors, vol. 21, no. 23, Article 8019, December 2021. doi: 10.3390/s21238019.
[22] J. Lv and Y. Pei, "Driver fatigue prediction method based on BiLSTM and novel comprehensive fatigue index," Human-Centric Intelligent Systems, vol. 5, pp. 545–558, March 2025. doi: 10.1007/s44230-025-00113-6.
[23] Z. Lian, T. Xu, Z. Yuan, J. Li, N. V. Thakor, and H. Wang, "Driving fatigue detection based on hybrid electroencephalography and eye tracking," IEEE Journal of Biomedical and Health Informatics, vol. 28, no. 11, pp. 6568–6580, November 2024. doi: 10.1109/JBHI.2024.3446952.
[24] W. Sun, Y. Wang, B. Hu, and Q. Wang, "Exploration of eye fatigue detection features and algorithm based on eye-tracking signal," Electronics, vol. 13, no. 10, Article 1798, May 2024. doi: 10.3390/electronics13101798.
[25] A. Kuwahara, K. Nishikawa, R. Hirakawa, H. Kawano, and Y. Nakatoh, "Eye fatigue estimation using blink detection based on Eye Aspect Ratio Mapping (EARM)," Cognitive Robotics, vol. 2, pp. 50–59, 2022. doi: 10.1016/j.cogr.2022.01.003.
Downloads
Published
Conference Proceedings Volume
Section
License
Copyright (c) 2026 Adroid Conference Series: Engineering and Technology
This work is licensed under a Creative Commons Attribution 4.0 International License.
