Demo Video

ICONIP 2012 Slides (English)

BS Defense Slides (Chinese)

Download Full Video
Computer Vision - Fuzzy Integral Model
Computer Vision - Runtime Feature Fusion

Fatigue driving is one of the major causes of traffic accidents. Additionally, the prediction of fatigue state is of crucial significance for pilots, drivers and chief conductors. Nevertheless, most of the existing Fatigue Driving Detection Systems (FDDS) are based on single signal source or extract single feature, which suffer from dim lighting, varying skin color, inaccuracy and high cost. Hence, this thesis is the first to propose a cheap, easy-to-use, accurate FDDS based on video signals, which combines geometry and texture features and predicts fatigue state by combining video with Electroencephalography(EOG) signals and gripping power. The author is the first to propose fatigue driving experiments by combining video and EOG signals on ICONIP 2012: EOG is more accurate while video includes more information; by combing both signals, we greatly improve the prediction accuracy. This system won the 2nd prize on ICCF 2011. Meanwhile, the author conducted similar experiments with gripping power, which again proves the robustness and accuracy.

Selected Reviews from ICONIP 2012

R1: The authors proposed a novel system to analysis vigilance level combining both video and Electrooculography (EOG) features. The new system overcomes some bars and limitations. In my view, it is a really nice result. I strongly recommend it to be published in the conference proceeding and possible special issue in journal after conference.
R2: A novel system for vigilance analysis based on both video and EOG features proposed in this paper. From the experimental results, we can see that their new method offers a good prediction of the actual error rate of the human being, which largely reflects the real vigilance level.
R3: This method outperforms the existing approaches using either video features or EOG features alone, since our proposed method utilizes both the accuracy of EOG signals but the yawn state and body postures provided by video as well.

Paper

Proceedings of the International Conference on Neural Information Processing (ICONIP), 2012.

Thesis

B.S. Thesis at Shanghai Jiao Tong University

Bachelor Defense

Jun 18, 2013 | Bachelor Defense in Shanghai Jiao Tong University | Shanghai, China

ICONIP 2012 Talk

Nov. 9, 2012 | ICONIP 2012 | Doha, Qatar

ACM Class Academic Festival Talk

Jul. 4, 2012 | KeyNote Lecture on the 3rd Academic Festival of ACM Class | Shanghai, China

BibTeX

@inproceedings{
    du2012online,
    title={Online Vigilance Analysis Combining Video and Electrooculography Features},
    author={Du, Ruofei and Liu, Renjie and Wu, Tianxiang and Lu, Bao-liang},
    booktitle={Neural Information Processing - 19th International Conference, {ICONIP} 2012},
    pages={447--454},
    year={2012},
    url={http://dx.doi.org/10.1007/978-3-642-34500-5_53},
    doi={10.1007/978-3-642-34500-5_53},
    publisher={Springer}
}

Citation

Du, R. F., Liu, R. J., Wu, T. X., & Lu, B. L. (2012, January). Online Vigilance Analysis Combining Video and Electrooculography Features. In Neural Information Processing - 19th International Conference, ICONIP 2012 (pp. 447-454). Springer Berlin Heidelberg.