DETECCIÓN DE SOMNOLENCIA EN CONDUCTORES DE VEHÍCULOS POR MEDIO DE PROCESAMIENTO DE VIDEO (DROWSINESS DETECTION IN VEHICLE DRIVERS THROUGH VIDEO PROCESSING)
Resumen
Los accidentes automovilísticos son una de las principales causas de muerte y lesiones a nivel mundial. Muchos son causados por fatiga y somnolencia de los conductores. El presente estudio tiene como objetivo detectar somnolencia en conductores de vehículos. La metodología del trabajo consistió en las siguientes etapas: en primer lugar, se empleó un algoritmo para la detección del rostro del sujeto dentro de la cabina de un automóvil durante la simulación de conducción para identificar regiones que incluyan cada ojo. Posteriormente se construyó un clasificador para distinguir las regiones de cada ojo como: abierto o cerrado. Finalmente, se desarrolló un algoritmo para el seguimiento de las regiones de interés para alimentar con imágenes al clasificador; para la detección se somnolencia se utiliza un criterio basado en una cantidad de fotogramas consecutivos presentando una identificación de ojos cerrados. El algoritmo presentó un 91.4% de exactitud en la detección de somnolencia.
Palabras Clave: Clasificación, somnolencia, inteligencia artificial.
Abstract
Car accidents are one of the leading causes of death and injury worldwide. Many are caused by fatigue and drowsiness in drivers. The present study is a work based on artificial intelligence applied to the field of video processing to determine a drowsy condition in vehicle drivers. The work methodology consisted of the following stages: first, using the first frame of the video, the subject's face is detected inside the car cabin during the driving simulation in order to identify the regions that include each eye. The next section consists of the construction of a classifier to identify the regions that each eye includes as: open eye or closed eye. Finally, an algorithm is developed to track the face and the regions of interest (regions that include the eyes) to feed the classifier with images; Detection of subject drowsiness is determined using criteria based on a series of consecutive frames having a closed eye identification. The algorithm presented a 91.4% accuracy in detecting drowsiness.
Keywords: Classification, drowsiness, artificial intelligence.
Texto completo:
700-721 PDFReferencias
Asthana, Akshay and Zafeiriou, Stefanos and Cheng, Shiyang and Pantic, Maja, Robust discriminative response map fitting with constrained local models, Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., pp. 3444–3451, 2013, doi: 10.1109/CVPR.2013.442.
Bacivarov, I., Ionita, M., and Corcoran, P., Statistical models of appearance for eye tracking and eye-blink detection and measurement. IEEE transactions on consumer electronics, 54(3), 1312-1320, 2008.
Celecia, A., Figueiredo, K., Vellasco, M., and González, R., A portable fuzzy driver drowsiness estimation system. Sensors, 20(15), 4093, 2020.
Chaabene, S., Bouaziz, B., Boudaya, A., Hökelmann, A., Ammar, A., and Chaari, L., Convolutional Neural Network for Drowsiness Detection Using EEG Signals. Sensors, 21(5), 1734, 2021.
Dunbar, J., Gilbert, J., and B. Lewis, B., Exploring differences between self-report and electrophysiological indices of drowsy driving: A usability examination of a personal brain-computer interface device, J. Safety Res., Apr. 2020, doi: 10.1016/j.jsr.2020.04.006.
Flores, M., Armingol, J., and De La Escalera, A., Sistema Avanzado de Asistencia a la Conducción para la Detección de la Somnolencia, Rev. Iberoam. Automática e Informática Ind., vol. 8, pp. 216–228, 2011, doi: 10.1016/j.riai.2011.06.009.
Hong, S., and Baek, H., Drowsiness Detection Based on Intelligent Systems with Nonlinear Features for Optimal Placement of Encephalogram Electrodes on the Cerebral Area. Sensors, 21(4), 1255, 2021.
Ji, Qiang and Yang, Xiaojie, Real-Time Eye, Gaze, and Face Pose Tracking for Monitoring Driver Vigilance, pp. 357–377, 2002, doi: 10.1006/rtim.2002.0279.
Kundinger, T., Sofra, N., and Riener, A., Assessment of the potential of wrist-worn wearable sensors for driver drowsiness detection. Sensors, 20(4), 1029, 2020.
Liu, Caifeng and Feng, Lin and Guo, Shuai and Wang, Huibing and Liu, Shenglan and Qiao, Hong, An incrementally cascaded broad learning framework to facial landmark tracking, Neurocomputing, vol. 410, pp. 125–137, Oct. 2020, doi: 10.1016/j.neucom.2020.05.025.
Liu, Qingshan and Yang, Jing and Deng, Jiankang and Zhang, Kaihua, Robust facial landmark tracking via cascade regression, Pattern Recognit., vol. 66, pp. 53–62, Jun. 2017, doi:10.1016/j.patcog.2016.12.024.
Lucas, Bruce D and Kanade, Takeo and others, Iterative Image Registration Technique With an Application To Stereo Vision., vol. 2, no. April 1981, pp. 674–679, 1981.
Maior, C., das Chagas Moura, M., Santana, J., and Lins, I., Real-time classification for autonomous drowsiness detection using eye aspect ratio. Expert Systems with Applications, 158, 113505, 2020.
MathWorks, Face Detection and Tracking Using the KLT Algorithm - MATLAB & Simulink - MathWorks América Latina. Disponible en: https://la.mathworks.com/help/vision/examples/face-detection-and-tracking-using-the-klt-algorithm.html (accessed Sep. 03, 2020).
N. Nojiri, X. Kong, L. Meng and H. Shimakawa, "Discussion on Machine Learning and Deep Learning based Makeup Considered Eye Status Recognition for Driver Drowsiness," Procedia, vol. 147, 2019, doi: 10.1016/j.procs.2019.01.252 , pp. 264-270.
Satti, A., Kim, J., Yi, E., Cho, H., and Cho, S., Microneedle array electrode-based wearable EMG system for detection of driver drowsiness through steering wheel grip. Sensors, 21(15), 5091, 2021.
Siddiqui, H. U. R., Saleem, A. A., Brown, R., Bademci, B., Lee, E., Rustam, F., & Dudley, S., Non-invasive driver drowsiness detection system. Sensors, 21(14), 4833, 2021.
Slater, Jeremy D, A definition of drowsiness: One purpose for sleep?, Med. Hypotheses, vol. 71, no. 5, pp. 641–644, Nov. 2008, doi: 10.1016/j.mehy.2008.05.035.
Vesselenyi, T., Moca, S., Rus, A., Mitran, T., and Tǎtaru, B., Driver drowsiness detection using ANN image processing, IOP Conf. Ser. Mater. Sci. Eng., vol. 252, no. 1, 2017, doi: 10.1088/1757-899X/252/1/012097.
Viola, P., and Jones, M., Robust Real-Time Face Detection, Int. J. Comput. Vis., vol. 57, no. 2, pp. 137–154, May 2004, doi: 10.1023/B:VISI.0000013087.49260.fb.
URL de la licencia: https://creativecommons.org/licenses/by/3.0/deed.es
Pistas Educativas está bajo la Licencia Creative Commons Atribución 3.0 No portada. 
TECNOLÓGICO NACIONAL DE MÉXICO / INSTITUTO TECNOLÓGICO DE CELAYA
Antonio García Cubas Pte #600 esq. Av. Tecnológico, Celaya, Gto. México
Tel. 461 61 17575 Ext 5450 y 5146
pistaseducativas@itcelaya.edu.mx