Citation

Sadeque, Md. Golam and Ahmmed, Tanvir and Hossain, Md. Aowal and Rahman, Md. Naimur and Sarker, Md Tanjil and Ramasamy, Gobbi (2025) A Novel CWT-CNN Framework for Driver Fatigue Detection from EEG Signals. In: 2025 Multimedia University Engineering Conference, MECON 2025, 21 July 2025 - 23 July 2025, Cyberjaya.

Text IEEE Xplore Full-Text PDF_8.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Fatigued driving has been a significant contributor to traffic accidents globally, posing serious threats to both human life and economic stability. Machine learning techniques based on electroencephalography (EEG) are showing promise for detecting driver fatigue, excelling other physiological modalities in this regard. However, it requires a lot of exertion, demands domain knowledge, and may not generalize effectively across different datasets to manually extract features from EEG signals. Therefore, investigating innovative deep-learning architectures that can effectively extract discriminative characteristics from unprocessed EEG data is necessary. This paper proposes a novel framework for driver fatigue detection from EEG signals using Continuous Wavelet Transform (CWT) and Convolutional Neural Networks (CNN). Rather than depending on hand-crafted feature extraction, we use CWT to convert EEG data into timefrequency spectrum pictures. After concatenating these spectral images from every channel, a CNN is fed to learn discriminative features for driver normal and fatigued states automatically. An average classification accuracy of 98.3% is obtained by evaluating the proposed CWT-CNN system on a publicly available EEG dataset comprising recordings from twelve subjects. The results obtained indicate that the CWT-CNN framework has enormous potential to create strong driver fatigue detection systems, thereby enhancing road safety.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Deep learning
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions:	Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	17 Mar 2026 02:31
Last Modified:	17 Mar 2026 02:31
URII:	http://shdl.mmu.edu.my/id/eprint/15463

Downloads

Edit (login required)