Distracted driving is the main cause for large number of motor vehicle accidents across the globe. Detecting a distracted driver is considered as the significant research area for reducing the road accidents. This paper focuses on a methodology to reduce the accidents caused by distracted driver with deep neural networks. CNN based method is used to develop the actions of driver from driver image dataset, which is used to classify the distracted driver into different categories. Proposed system consist of three models namely, vanilla CNN, vanilla CNN with data augmentation, and CNN with transfer learning. Deep neural network model is developed from a state farm dataset, which consists of 10 actions in 26 different subjects such as texting, mobile phone usage in driving, delayed arrival, normal driving, alcohol consumption etc. Results obtained from 5 Epochs shows that all the experiments have exceeded 75% accuracy and the best observed result is 97%.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Real Time Detection of driver distraction using CNN


    Contributors:


    Publication date :

    2020-08-01


    Size :

    1472043 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English





    DRIVER DISTRACTION DETECTION

    HERMAN DAVID MICHAEL | European Patent Office | 2021

    Free access