Máquinas de soporte vectorial para inferir el punto de atención de automovilistas vistiendo lentes inteligentes

Most of the methods for inferring distraction while driving are based on visual characteristics of head posture, since it is a strong indicator of distraction while driving. This paper proposes the use of mounted inertial sensors on intelligent lenses. Hence, data was collected from five participants and experiments were conducted to evaluate the feasibility of using support vector machines (SVM) to generate models of drivers to infer the point of attention. Results show an acceptable performance of the SVM to identify specific positions of the car cabin on which drivers focus their attention. Until now, an average accuracy of 83.42% has been achieved.

[1]  Mohan M. Trivedi,et al.  Head Pose Estimation and Augmented Reality Tracking: An Integrated System and Evaluation for Monitoring Driver Awareness , 2010, IEEE Transactions on Intelligent Transportation Systems.

[2]  Brian N. Fildes,et al.  Review of crash effectiveness of intelligent transport systems , 2007 .

[3]  Hang-Bong Kang,et al.  Various Approaches for Driver and Driving Behavior Monitoring: A Review , 2013, 2013 IEEE International Conference on Computer Vision Workshops.

[4]  G. G. Ray,et al.  Human Factors in Safe Driving - A Review of Literature on Systems Perspective, Distractions and Errors , 2012, 2012 IEEE Global Humanitarian Technology Conference.

[5]  Luís A. Castro,et al.  Support Vector Machines for Inferring Distracted Behavior of Drivers Wearing Smart Glasses , 2016, UCAmI.

[6]  David J. Spiegelhalter,et al.  Machine Learning, Neural and Statistical Classification , 2009 .

[7]  Jinhui Tang,et al.  Real-Time System for Driver Fatigue Detection by RGB-D Camera , 2015, ACM Trans. Intell. Syst. Technol..

[8]  Yi-Chun Lin,et al.  Driver distraction recognition based on dual compass motion sensoring , 2014, 17th International IEEE Conference on Intelligent Transportation Systems (ITSC).

[9]  Thomas A. Dingus,et al.  The Impact of Driver Inattention on Near-Crash/Crash Risk: An Analysis Using the 100-Car Naturalistic Driving Study Data , 2006 .

[10]  Bernt Schiele,et al.  A tutorial on human activity recognition using body-worn inertial sensors , 2014, CSUR.

[11]  Chun-Fu Chuang,et al.  HMM-based driving behavior recognition for in-car control service , 2015, 2015 IEEE International Conference on Consumer Electronics - Taiwan.

[12]  Jamie Chamberlin.  Smart glasses: Driver distraction or safety tool?: (505762014-005) , 2014 .

[13]  Bao-Liang Lu,et al.  A novel approach to driving fatigue detection using forehead EOG , 2015, 2015 7th International IEEE/EMBS Conference on Neural Engineering (NER).

[14]  Özlem Durmaz Incel.  Analysis of Movement, Orientation and Rotation-Based Sensing for Phone Placement Recognition , 2015, Sensors.

[15]  Luís A. Castro,et al.  Reducing Drivers' Distractions in Phone-Based Navigation Assistants Using Landmarks , 2013, UCAmI.

[16]  Ameet Talwalkar,et al.  Foundations of Machine Learning , 2012, Adaptive computation and machine learning.

[17]  Mohan M. Trivedi,et al.  Robust and continuous estimation of driver gaze zone by dynamic analysis of multiple face videos , 2014, 2014 IEEE Intelligent Vehicles Symposium Proceedings.