In today’s day and age, road accidents attributed to drowsy driving and reckless overtaking are escalating for various reasons. This research paper introduces a comprehensive and dynamic real-time driver safety system that harnesses advanced deep learning algorithms and sensor-based technology. The research is divided into three systems that act as a guardian, meant to detect any such drowsiness, or angry behaviour of the driver and makes sure both the driver and the vehicles behind are made aware of the situation, thereby mitigating potential accidents. Such a system requires continuous monitoring of driver behaviour, taking into account metrics like eye aspect ratio (EAR) and lip distance analysis. The research delves into overtaking detection, which is done by integrating ultrasonic sensors, the Collision Prediction System measures the proximity between the vehicle and the other vehicles ahead of it and alerts the driver and honks when the distance is at very close proximity. Our system uses a camera to monitor the driver’s facial expression and issues warnings through the rear alert display and interior lighting cues if drowsiness or anger is detected. If extreme drowsiness is detected, immediate intervention is triggered through visual and auditory cues. The Overtake Safety Awareness System makes use of the front cameras and gives a view of the subject vehicle to the rear vehicles. Experimentation of the system showcases an overtaking vehicle detection accuracy exceeding 90%, signifying its robustness in identifying critical driving scenarios. Through its innovative blend of advanced sensors, intelligent algorithms and driver-centric design this research paper talks about the implementation of a real-time driver safety system that adeptly identifies and prevents drowsy driving and reckless overtaking.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Advancing Road Safety: Deep Learning-Powered Real-Time Driver State Assessment and R-CNN for Proximity Vehicle Monitoring


    Weitere Titelangaben:

    Lect. Notes in Networks, Syst.


    Beteiligte:
    Swaroop, Abhishek (Herausgeber:in) / Kansal, Vineet (Herausgeber:in) / Fortino, Giancarlo (Herausgeber:in) / Hassanien, Aboul Ella (Herausgeber:in) / Mallegowda, M. (Autor:in) / Kumaran, Shubeeksh (Autor:in) / Aditya Raj, V. (Autor:in) / Kumar, Skanda S. (Autor:in) / Gowda, Ronith H. (Autor:in)

    Kongress:

    Doctoral Symposium on Computational Intelligence ; 2024 ; Lucknow, India May 10, 2024 - May 10, 2024



    Erscheinungsdatum :

    04.10.2024


    Format / Umfang :

    15 pages





    Medientyp :

    Aufsatz/Kapitel (Buch)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch





    Deep Learning Based Real-Time Driver Emotion Monitoring

    Verma, Bindu / Choudhary, Ayesha | IEEE | 2018


    ACTIVE SAFETY DRIVER ASSISTANCE SYSTEM BASED ON DRIVER REAL-TIME STATE MONITORING

    LIU CAIXIA | Europäisches Patentamt | 2024

    Freier Zugriff

    Deep Learning for Driver Vigilance and Road Safety

    Kaushik, Priyanka / Yadav, Khushbu / C, Sincija et al. | IEEE | 2024


    A Real-Time Machine Learning-Based Road Safety Monitoring and Assessment System

    Fowdur, Tulsi Pawan / Hawseea, Mohammed Fayez | Springer Verlag | 2024