Detection of Distracted Driver Using Convolutional Neural Network

Number of road accidents is continuously increasing in last few years worldwide. As per the survey of National Highway Traffic Safety Administrator, nearly one in five motor vehicle crashes are caused by distracted driver. We attempt to develop an accurate and robust system for detecting distracted driver and warn him against it. Motivated by the performance of Convolutional Neural Networks in computer vision, we present a CNN based system that not only detects the distracted driver but also identifies the cause of distraction. VGG-16 architecture is modified for this particular task and various regularization techniques are implied in order to improve the performance. Experimental results show that our system outperforms earlier methods in literature achieving an accuracy of 96.31% and processes 42 images per second on GPU. We also study the effect of dropout, L2 regularization and batch normalisation on the performance of the system. Next, we present a modified version of our architecture that achieves 95.54% classification accuracy with the number of parameters reduced from 140M in original VGG-16 to 15M only.

[1]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Mohan M. Trivedi,et al.  In-vehicle hand activity recognition using integration of regions , 2013, 2013 IEEE Intelligent Vehicles Symposium (IV).

[3]  A. Ng.  Feature selection, L1 vs. L2 regularization, and rotational invariance , 2004, Twenty-first international conference on Machine learning - ICML '04.

[4]  Bailing Zhang,et al.  Erratum to: Recognition of driving postures by combined features and random subspace ensemble of multilayer perceptron classifiers , 2012, Neural Computing and Applications.

[5]  Andreas Savakis,et al.  Distracted Driver Detection: Deep Learning vs Handcrafted Features , 2017 .

[6]  Nanning Zheng,et al.  Visual recognition of driver hand-held cell phone use based on hidden CRF , 2011, Proceedings of 2011 IEEE International Conference on Vehicular Electronics and Safety.

[7]  Mohan M. Trivedi,et al.  On Performance Evaluation of Driver Hand Detection Algorithms: Challenges, Dataset, and Metrics , 2015, 2015 IEEE 18th International Conference on Intelligent Transportation Systems.

[8]  Mohan M. Trivedi,et al.  Driver hand activity analysis in naturalistic driving studies: challenges, algorithms, and experimental studies , 2013, J. Electronic Imaging.

[9]  Hesham M. Eraqi,et al.  Real-time Distracted Driver Posture Classification , 2017, ArXiv.

[10]  Frans Coenen,et al.  Driving posture recognition by convolutional neural networks , 2015, 2015 11th International Conference on Natural Computation (ICNC).

[11]  Marios Savvides,et al.  Driver cell phone usage detection on Strategic Highway Research Program (SHRP2) face view videos , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[12]  Marios Savvides,et al.  Multiple Scale Faster-RCNN Approach to Driver’s Cell-Phone Usage and Hands on Steering Wheel Detection , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[13]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[14]  Bailing Zhang,et al.  Recognition of driving postures by contourlet transform and random forests , 2012 .

[15]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[16]  Mohan M. Trivedi,et al.  Understanding head and hand activities and coordination in naturalistic driving videos , 2014, 2014 IEEE Intelligent Vehicles Symposium Proceedings.

[17]  Mohan M. Trivedi,et al.  Head, Eye, and Hand Patterns for Driver Activity Recognition , 2014, 2014 22nd International Conference on Pattern Recognition.