The paper talks about Quantification of Alertness for vision based Driver Drowsiness and Alertness Warning System (DDAWS). The quantification of alertness, as per Karolinska Sleepiness Scale (KSS), reads the basic input of facial features & behaviour recognition of driver in a standard manner. Although quantification of alertness is inconclusive with respect to the true value, the paper emphasised on systematic validation process of the system covering various scenarios in order to evaluate the system’s functionality very close to the reality. The methodology depends on definition of threshold values of blink and head pose. The facial features are defined by number of blinks with classification of heavy blink and light blink and head pose in (x, y, z) directions. The Human Machine Interface (HMI) warnings are selected in the form of visual and acoustic signals. Frequency, Amplitude and Illumination of HMI alerts are specified. The protocols and trigger functions are defined and KSS stage is calculated for selected duration of time and frames of data covering demographics of people, road types, weather conditions and human behavioural actions. Multiple iterations of threshold values are conducted for test, the outcomes are listed and results are analysed. In order to enhance the system’s robustness and reliability of the obtained results, the paper added provision of secondary strategies based on vehicle metrics like Standard Lane Deviation Laterally (SDLAT), Yaw rate, latest activation of controls, reversal steering correction rate etc. The paper concludes with the threshold value of eye blink, HMI warning, results of hybrid approach of driver drowsiness and alertness warning system and discussed the scope of improvement.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Quantification of Alertness and Evaluation Method for Vision Based Driver Drowsiness and Alertness Warning System


    Additional title:

    Sae Technical Papers



    Conference:

    Symposium on International Automotive Technology ; 2024



    Publication date :

    2024-01-16




    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    English




    DRIVER ALERTNESS WARNING SYSTEM AND METHOD

    HUBBARD JONATHAN / KANTARJIEV CHRISTOPHER A | European Patent Office | 2021

    Free access

    DRIVER ALERTNESS WARNING SYSTEM AND METHOD

    HUBBARD JONATHAN / KANTARJIEV CHRISTOPHER A | European Patent Office | 2019

    Free access

    DRIVER ALERTNESS WARNING SYSTEM AND METHOD

    HUBBARD JONATHAN / KANTARJIEV CHRISTOPHER A | European Patent Office | 2024

    Free access

    Driver alertness warning system and method

    HUBBARD JONATHAN / KANTARJIEV CHRISTOPHER A | European Patent Office | 2020

    Free access

    Driver alertness warning system and method

    HUBBARD JONATHAN / KANTARJIEV CHRISTOPHER A | European Patent Office | 2024

    Free access