Comprehensive Benchmark of Yolov11n, SSD MobileNet, CenterFace, Yunet, FastMtCnn, HaarCascade, and LBP for Face Detection in Video Based Driver Drowsiness
- Agnestia Agustine Djoenaidi Go Universitas Dian Nuswantoro, Semarang, Indonesia
-
Farrikh Alzami
*
Universitas Dian Nuswantoro, Semarang,
Indonesia
- Muhammad Naufal Universitas Dian Nuswantoro, Semarang, Indonesia
- Harun Al Azies Universitas Dian Nuswantoro, Semarang, Indonesia
- Sri Winarno Universitas Dian Nuswantoro, Semarang, Indonesia
- Ricardus Anggi Pramunendar Universitas Dian Nuswantoro, Semarang, Indonesia
- Rama Aria Megantara Universitas Dian Nuswantoro, Semarang, Indonesia
- Isa Iant Maulana Universitas Dian Nuswantoro, Semarang, Indonesia
- Mohammad Arif Universitas Dian Nuswantoro, Semarang, Indonesia (*) Corresponding Author
Abstract
Face detection is a critical foundation of video-based drowsiness monitoring systems because all downstream tasks such as eye-closure estimation, yawning detection, and head movement analysis depend entirely on correctly identifying the face region. Many previous studies rely on detector-generated outputs as ground truth, which can introduce bias and inflate model performance . To avoid this limitation, I manually constructed a ground truth dataset using 1,229 frames extracted from 129 yawning and microsleep videos in the NITYMED dataset. Ten representative frames were sampled from each video using a face-guided extraction script, and all frames were manually annotated in Roboflow following the COCO format to ensure accurate bounding box labeling under varying lighting, head poses, and facial deformation. Using this manually annotated dataset, I conducted a comprehensive benchmark of seven face-detection algorithms: YOLOv11n, SSD MobileNet, CenterFace, YuNet, FastMtCnn, HaarCascade, and LBP. The evaluation focused on localization quality using Intersection over Union (IoU ≥ 0.5) and Dice Similarity, allowing each algorithm’s predicted bounding box to be directly compared against human defined ground truth. The results show that HaarCascade achieved the highest IoU and Dice scores, particularly in frontal and well-lit frames. FastMtCnn also produced strong alignment with a high number of correctly matched frames. CenterFace and SSD MobileNet demonstrated smooth bounding box fitting with competitive Dice scores, while YOLOv11n and YuNet delivered moderate but stable performance across most samples. LBP showed the weakest results, mainly due to its sensitivity to lighting variations and soft-texture regions. Overall, this benchmark provides an unbiased and comprehensive comparison of modern and classical face-detection algorithms for video-based driver-drowsiness applications.
Downloads
References
Nur Rachmi Widyastuti and Dani Fitria Brilianti, “Impact of Drowsiness on Road Traffic Accidents in Yogyakarta,” Journal of Scientific Research, Education, and Technology (JSRET), vol. 3, no. 4, pp. 1651–1661, 2024, doi: 10.58526/jsret.v3i4.555.
Farrikh Alzami, Muhammad Naufal, Harun Al Azies, Sri Winarno, and Moch Arief Soeleman, “Time Distributed MobileNetV2 with Auto-CLAHE for Eye Region Drowsiness Detection in Low Light Conditions,” (IJACSA) International Journal of Advanced Computer Science and Applications, p. 13, 2024, doi: 10.14569/IJACSA.2024.0151146.
Anna W. T. Cai, Jessica E. Manousakis, and Bikram Singh, “On‑road driving impairment following sleep deprivation differs according to age,” Scientific Reports, vol. 11, p. 21561, 2021, doi: 10.1038/s41598-021-99133-y.
Siham Essahraui, Ismail Lamaakal, and Ikhlas El Hamly, “Real-Time Driver Drowsiness Detection Using Facial Analysis and Machine Learning Techniques,” Sensors, vol. 25, no. 3, p. 812, 2025, doi: 10.3390/s25030812.
Shehzad Saleem, “Risk Assessment of Road Traffic Accidents Related to Sleepiness During Driving: A Systematic Review,” East Mediterranean Health Journal, vol. 28, no. 9, pp. 695–700, 2022, doi: 10.26719/emhj.22.055.
Adetayo Olugbenga Onososen, Innocent Musonda, and Damilola Onatayo, “Drowsiness Detection of Construction Workers: Accident Prevention Leveraging YOLOv8 Deep Learning and Computer Vision Techniques,” Buildings, vol. 15, no. 3, p. 500, 2025, doi: 10.3390/buildings15030500.
Ramadan TH. Hasan and Amira Bibo Sallow, “Face Detection and Recognition Using OpenCV,” JOURNAL OF SOFT COMPUTING AND DATA MINING, vol. 2, no. 2, p. 12, 2021, doi: https://doi.org/10.30880/jscdm.2021.02.02.008.
Guodong Guo and Na Zhang, “A survey on deep learning based face recognition,” Computer Vision and Image Understanding, vol. 189, p. 102805, 2019, doi: https://doi.org/10.1016/j.cviu.2019.102805.
Z. Cai, K. Zhou, and Z. Liao, “A Systematic Review of YOLO-Based Object Detection in Medical Imaging: Advances, Challenges, and Future Directions,” Computers, Materials and Continua, vol. 85, no. 2, pp. 2255–2303, Sep. 2025, doi: 10.32604/cmc.2025.067994.
Yilin Liu, Ruian Liu, Shengxiong Wang, Da Yan, Bo Peng, and Tong Zhang, “Video Face Detection Based on Improved SSD Model and Target Tracking Algorithm,” Journal of Web Engineering, vol. 21, no. 2, 2022, doi: https://doi.org/10.13052/jwe1540-9589.21225.
C. Gheorghe, M. Duguleana, R. G. Boboc, and C. C. Postelnicu, “Analyzing Real-Time Object Detection with YOLO Algorithm in Automotive Applications: A Review,” CMES - Computer Modeling in Engineering and Sciences, vol. 141, no. 3, pp. 1939–1981, Oct. 2024, doi: 10.32604/cmes.2024.054735.
Anurag Pandey, Divyansh Choudhary, Ritik Agarwal, Tushar Shrivastava, and Kriti, “Face detection using Haar cascade classifier,” AECE 2022, p. 599, 2022, doi: 10.2139/ssrn.4157631.
Hruthik S. Upendra, Shruti Suman, Sai S. Vishnu, and Jaya Dharani, “Real-Time Face Mask Detection using OpenCV and Deep Learning,” CEUR-WS (Vol. 3085), p. 6, Sep. 2021, doi: 10.1109/AECE62803.2024.10911331.
Adamu Ali-Gombe, Eyad Elyan, Carlos Francisco Moreno-García, and Johan Zwiegelaar, “Face Detection with YOLO on Edge,” in Proceedings of the International Neural Networks Society (INNS, volume 3). Springer, Cham, 2021, pp. 284–292. doi: 10.1007/978-3-030-80946-1_25.
Noor Afiza Binti Mohd Ariffin, Usman Abdul Gimba, and Ahmad Musa, “Face detection based on Haar Cascade and Convolution Neural Network (CNN),” Journal of Advanced Research in Computing and Applications, vol. 38, p. 11, 2025, doi: https://doi.org/10.37934/arca.38.1.111.
Gaiping Liu, Jianmei Xiao, and Xihuai Wang, “Optimization of Face Detection Algorithm based on MTCNN,” Semantic, [Online]. Available: https://www.semanticscholar.org/paper/Optimization-of-Face-Detection-Algorithm-based-on-Liu-Xiao/8e23ccf923cb9223accf282582919817c89b0706
Olena Yakovleva, Andrii Kovtunenko, Valentyn Liubchenko, Vadym Honcharenko, and Oleg Kobylin, “Face Detection for Video Surveillance-based Security System”, International Conference on Computational Linguistics and Intelligent Systems, April, 2023
Radimas Putra M.D.L, Sirojul Hadi, and Parama Diptya Widayaka, “Low Cost System for Face Mask Detection Based Haar Cascade Classifier Method”, Matrik, vol. 21, no. 1, 2021, doi: 10.30812/matrik.v21i1.1187.
Yap Jia Hui and Lee Siaw Chong, “Face Detection withHaar Cascades Method,” Enhanced Knowledge in Sciences and Technology, vol. 2, no.1, 2022, doi: https://doi.org/10.30880/ekst.2022.02.01.033.
Adlan Hakim Ahmad et al., “Real time face recognition of video surveillance system using haar cascade classifier” Indonesian Journal of Electrical Engineering and Computer Science (IJEECS) vol. 21, no. 3, 2021, doi: :10.11591/ijeecs.v21.i3.pp1389-1399.
Wei Wu, Hanyang Peng, and Shiqi Yu, “YuNet: A Tiny Millisecond-level Face Detector,” Machine Intelligence Research, vol. 20, April, 2023, doi: 10.1007/s11633-023-1423-y.
Bila bermanfaat silahkan share artikel ini
Berikan Komentar Anda terhadap artikel Comprehensive Benchmark of Yolov11n, SSD MobileNet, CenterFace, Yunet, FastMtCnn, HaarCascade, and LBP for Face Detection in Video Based Driver Drowsiness
Pages: 1775-1784
Copyright (c) 2025 Agnestia Agustine Djoenaidi Go, Farrikh Alzami, Muhammad Naufal, Harun Al Azies, Sri Winarno, Ricardus Anggi Pramunendar, Rama Aria Megantara, Isa Iant Maulana, Mohammad Arif
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under Creative Commons Attribution 4.0 International License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (Refer to The Effect of Open Access).
