The traffic accident prediction represents a vital role in the combined planning and management of traffic, the design which involve some nonlinear components, such as people, road, vehicle, weather and so on. The convention way of extended studies cannot expose the real situation since the noise corruption and amount of data are too small, so that the result of prediction cannot be satisfied. This paper proposes deep learning-based new road traffic accident prediction applying a “Convolutional Neural Network model” (CNN). It uses traffic accident influencing circumstances like light, weather, traffic flow to make a state matrix describing the traffic state and CNN model. This paper utilizes examples to examine the accuracy of the proposed model. The empirical results show that the proposed model is more efficient than the current neural network design to predict traffic accidents compared with the traditional “Backpropagation” (BP) algorithm


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Traffic Accident Prediction based on CNN Model


    Contributors:


    Publication date :

    2021-05-06


    Size :

    2053249 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Training method of traffic accident prediction model and traffic accident prediction method and device

    WANG NIANMING / CHEN YANG / ZHOU MINGKE et al. | European Patent Office | 2024

    Free access

    Tunnel traffic flow and accident prediction model

    ZHENG QI / LI ZHIYANG / WANG PENGHUI | European Patent Office | 2025

    Free access

    Prediction system for traffic accident

    BACK JU YONG | European Patent Office | 2019

    Free access

    Road traffic accident prediction method

    WANG SHUNSHUN / YAN CHANGSHUN / SHAO YONG | European Patent Office | 2023

    Free access

    Prediction system for traffic accident

    BACK JU YONG | European Patent Office | 2019

    Free access