While urban settings offer interesting challenges, they present the need for quick identification of emergency response units to maintain order on the road and to safeguard citizens. This paper outlines a two-in-one system based on Convolution Neural Networks for visual recognition of emergency vehicles and Logistic Regression for auditory detection of sirens. It is made up of two components. The first one deals with the increase of its capability through different data preprocessing techniques. More specifically, a CNN model is built to classify a dataset of images containing ambulances, fire trucks, police cars, etc. and normal vehicle imagery. Local analysis is also achieved using components of the ‘Logistic Regression’ model, which helps in analyzing the traffic noise and classifying different sounding emergency vehicle sirens. The second component of the system is about sounds, which is solved using the method of Logistic Regression for recognizing calling sirens and separating them from the background traffic. Sound level patterns are stored in the processor which act as a template for differentiating the level of sirens and other vehicular sounds. The main objective of the system is to combine visual and audio detection techniques for a better speed of emergency response and traffic control especially when such emergencies occur in busy town centers. The combination of these two detection techniques offers a better approach to finding emergency vehicles especially with the growth of smart transport systems. This work indicates the favorable conditions under which delay systems and deep learning can be efficiently employed to improve safety as well as efficiency in traffic activities.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Smart Emergency Vehicle Detection: Merging Vision and Sound


    Beteiligte:
    Senthilselvi, A. (Autor:in) / J, Keshav (Autor:in) / Abraham, Jason (Autor:in) / Srinitish, R (Autor:in) / A, Yaduraj (Autor:in) / Pandi, S. Senthil (Autor:in)


    Erscheinungsdatum :

    20.12.2024


    Format / Umfang :

    832723 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Emergency Vehicle Detection using Vehicle Sound Classification: A Deep Learning Approach

    Sathruhan, S. / Herath, Oshadhi K. / Sivakumar, T. et al. | IEEE | 2022


    Emergency Vehicle Detection Using Deep ConvNet2D and Computer Vision

    Poongavanam, Elumalaivasan / A, Prathik. / T, Munirathinam. et al. | IEEE | 2024


    Vehicle Density and Emergency Vehicle Detection for Smart Traffic Control

    Vigneash, L. / Samraj, S. / Bhuvaneswari, S. et al. | IEEE | 2023


    Leveraging Computer Vision for Emergency Vehicle Detection-Implementation and Analysis

    Kaushik, S / Raman, Abhishek / Rajeswara Rao, K.V.S | IEEE | 2020


    Vehicle merging control system and vehicle merging control method, and failure detection device for vehicle control system

    TAKAHIRO IKEDA / KEIICHI KATSUTA / KENJI IMAMOTO | Europäisches Patentamt | 2017

    Freier Zugriff