Developing a Drowsiness Detection System for Safe Driving Using YOLOv9
Abstract
Drowsiness detection systems play a crucial role in safe driving, considering the high rate of traffic accidents caused mainly by drowsiness. Several drowsiness detection systems built using the eye aspect ratio (EAR), percentage of eyelid closure (PERCLOS), and convolutional neural network (CNN) methods still have limitations in terms of accuracy and response time. This study aimed to overcome these problems by applying the You Only Look Once version 9 (YOLOv9). This method has advantages in terms of speed and accuracy because it can detect objects in real-time in one processing stage. The dataset was collected independently from several sources in a real environment inside the vehicle with various lighting and viewing angles; then, labeling, preprocessing, and modeling processes were conducted. The model performance was evaluated based on precision, recall, F1 score, and mean average precision (mAP) metrics. The best model was optimized using several optimization techniques to determine the most optimal results. The results indicate that the YOLOv9 model trained using Nadam (Nesterov-accelerated adaptive moment estimation) optimization has a better image processing speed than other models. This model yielded a precision level of 99.4%, recall of 99.6%, F1 score of 99.5%, mAP@50 of 99.5%, mAP@50-95 of 85.5%, and a processing speed of 52.08 FPS. The developed model can detect drivers’ drowsiness signs, such as closed eyes, yawning, abnormal head positions, and unnatural hand movements, in real time. However, this model still has limitations in detecting drivers wearing sunglasses, so further development is needed to improve its performance in these conditions.
References
W.U. Puspoprodjo and N.N. Laila, “Studi pemahaman dan perilaku keselamatan berkendara (safety riding) pada remaja dan usia produktif di Pulau Jawa,” J. Ilm. Kesehat., vol. 20, no. 3, pp. 118–126, Dec. 2021, doi: 10.33221/jikes.v20i3.1480.
A.A.M. Suradi et al., “Sistem deteksi kantuk pengemudi mobil berdasarkan analisis rasio mata menggunakan computer vision,” JUKI J. Komput. Inform., vol. 5, no. 2, pp. 222–230, Nov. 2023, doi: 10.53842/juki.v5i2.269.
National Center for Statistics and Analysis, “Overview of motor vehicle traffic crashes in 2022,” Nat. Highw. Traffic Saf. Adm., Washington, D.C., USA, DOT HS 813 560, 2024.
R. Rahmadiyani and A. Widyanti, “Prevalence of drowsy driving and modeling its intention: An Indonesian case study,” Transp. Res. Interdisciplinary Perspect., vol. 19, May 2023, Art. no 100824, doi: 10.1016/j.trip.2023.100824.
S.S. Soleimanloo et al., “The association of schedule characteristics of heavy vehicle drivers with continuous eye-blink parameters of drowsiness,” Transp. Res. Part F, Traffic Psychol. Behaviour, vol. 90, pp. 485–499, Oct. 2022, doi: https://doi.org/10.1016/j.trf.2022.08.019.
B. Subekti, “Penanganan kecelakaan lalu lintas dalam perspektif kebijakan publik,” Indones. J. Edu. Humanit., vol. 2, no. 4, pp. 144–151, Nov. 2022.
R.S. Kamila, “Faktor yang berhubungan dengan kelelahan kerja pada pengemudi bus di PT X tahun 2022,” Undergraduate thesis, UIN Syarif Hidayatullah, Jakarta, Indonesia, 2022.
R. Aprianto, A. Rokhim, A. Basuki, and S. Sugiyarto, “Pengaruh karakteristik pengemudi dan pemanfaatan rest area terhadap kelelahan pengemudi studi kasus Ruas Jalan Tol Pejagan-Solo,” J. Keselam. Transp. Jalan (Indones. J. Road Saf.), vol. 8, no. 1, pp. 92–103, Jun. 2021, doi: 10.46447/ktj.v8i1.310.
M.Y. Petra and L. Lina, “Sistem pendeteksi kantuk pada mata pengendara roda empat dengan algoritma convolutional neural network,” Computatio J. Comp. Sci. Inf. Syst., vol. 8, no. 1, pp. 20–28, Apr. 2024, doi: 10.24912/computatio.v8i1.16220.
R. Rahmania et al., “Exploration of the impact of kernel size for YOLOv5-based object detection on quadcopter,” JOIV, Int. J. Inform. Visualization, vol. 6, no. 3, pp. 726–735, Sep. 2022, doi: 10.30630/joiv.6.3.898.
B.G. Amira, M.M. Zoulikha, and P. Hector, “Driver drowsiness detection and tracking based on YOLO with Haar cascades and ERNN,” Int. J. Saf. Secur. Eng., vol. 11, no. 1, pp. 35–42, Feb. 2021, doi: 10.18280/ijsse.110104.
A.Y. Alin, K. Kusrini, and K.A. Yuana, “The effect of data augmentation in deep learning with drone object detection,” IJCCS (Indones. J. Comput. Cybern. Syst.), vol. 17, no. 3, pp. 237–248, Jul. 2023, doi: 10.22146/ijccs.84785.
R. Thakur, Shivam, S. Raj, and S. Pandey, “Driver drowsiness detection system using machine learning,” in Advances in Transdisciplinary Engineering, R.M. Singari, P.K. Kankar, Eds., Amsterdam, Netherlands: IOS Press BV, 2022, pp. 31–38.
H.V. Chand and J. Karthikeyan, “CNN based driver drowsiness detection system using emotion analysis,” Intell. Automat. Soft Comput., vol. 31, no. 2, pp. 717–728, 2022, doi: 10.32604/iasc.2022.020008.
A. Debsi et al., “Driver distraction and fatigue detection in images using ME-YOLOv8 algorithm,” IET Intell. Transp. Syst., vol. 18, no. 10, pp. 1910–1930, Oct. 2024, doi: 10.1049/itr2.12560.
J. Surya and F.H. Aminuddin, Pemrograman MYSQL Database With Streamlit Python. Kota Jambi, Jambi: PT. Sonpedia Publishing Indonesia, 2024.
D.N. Alfarizi et al., “Penggunaan metode YOLO pada deteksi objek: Sebuah tinjauan literatur sistematis,” AI SPK, J. Artif. Intel. Sist. Penunjang Keputusan, vol. 1, no. 1, pp. 54–63, Jun. 2023.
A. Nurhopipah, I.R. Azziz, and J. Suhaman, “Behind the mask: Detection and recognition based-on deep learning,” IJCCS (Indones. J. Comput. Cybern. Syst.), vol. 16, no. 1, pp. 67–78, Jan. 2022, doi: 10.22146/ijccs.72075.
W.A. Triyanto, K. Adi, and J.E. Suseno, “Detection and tracking of broiler flock movements in the chicken coop using YOLO,” in 8th Int. Conf. Energy Environ. Epidemiol. Inf. Syst. (ICENIS 2023), 2023, pp. 1–13.
M.T.A. Dipu, S.S. Hossain, Y. Arafat, and F.B. Rafiq, “Real-time driver drowsiness detection using deep learning,” Int. J. Adv. Comput. Sci. Appl., vol. 12, no. 7, pp. 844–850, Jul. 2021, doi: 10.14569/IJACSA.2021.0120794.
Ultralytics, “YOLOv9: A Leap Forward in Object Detection Technology.” Access date: . [Online]. Available: https://docs.ultralytics.com/models/yolov9/
M.F. Ridho, F. Panca, W. Yandi, and A.A. Rachmani, “Drowsiness detection in the advanced driver-assistance system using YOLO V5 detection model,” Electron, J. Ilm. Tek. Elekt., vol. 5, no. 1, pp. 130–137, May 2024, doi: 10.33019/electron.v5i1.136.
Z. Wang, K. Yao, and F. Guo, “Driver attention detection based on improved YOLOv5,” Appl. Sci., vol. 13, no. 11, Jun.-1 2023, Art. no 6645, doi: 10.3390/app13116645.
S. Liu et al., “A driver fatigue detection algorithm based on dynamic tracking of small facial targets using YOLOv7,” IEICE Trans. Inf. Syst., vol. E106-D, no. 11, pp. 1881–1890, 2023, doi: 10.1587/transinf.2023EDP7093.
Y. Du, X. Liu, Y. Yi, and K. Wei, “Optimizing road safety: Advancements in lightweight YOLOv8 models and GhostC2f design for real-time distracted driving detection,” Sensors, vol. 23, no. 21, Nov.-1 2023, Art. no 8844, doi: 10.3390/s23218844.
A.A. Permana, M.R. Muttaqin, and M. A. Sunandar, “Sistem deteksi api secara real time menggunakan algoritma You Only Look Once (YOLO) versi 8,” Jati (J. Mhs. Tek. Inform), vol. 8, no. 5, pp. 10395–10400, Oct. 2024, doi: 10.36040/jati.v8i5.10847.
© Jurnal Nasional Teknik Elektro dan Teknologi Informasi, under the terms of the Creative Commons Attribution-ShareAlike 4.0 International License.
1.png)
