The invention relates to a multi-modal highway congestion identification method and system. The method comprises the steps that A, a local client preprocesses collected highway multi-modal data; b, on the basis of the multi-modal data, the local client constructs a training model for various modal data; for each training model, the local client and the credible central server carry out collaborative iteration on the training model by utilizing federated learning and homomorphic encryption technologies, and an output model is generated after training is completed; c, performing back-end fusion of multi-modal data on the output model, namely fusing classifier output decisions trained by different modal data to obtain an average congestion degree output by the model; and D, based on the average congestion degree, optimizing and updating the congestion calculation interval of the model, and realizing adaptive correction of algorithm parameters of highway network congestion identification. According to the method and the system, the speed and the accuracy of highway congestion identification can be improved, and the security of data sharing is ensured.

    本发明涉及一种多模态高速公路拥堵识别方法及系统,该方法包括:A、本地客户端对收集的高速公路多模态数据进行预处理;B、基于多模态数据,本地客户端对各种模态数据构造训练模型;对于各个训练模型,本地客户端和可信的中央服务器利用联邦学习和同态加密技术进行协同迭代训练模型,训练完成后产生输出模型;C、对输出模型进行多模态数据的后端融合,即对不同模态数据分别训练好的分类器输出决策进行融合,得到模型输出的平均拥堵程度;D、基于平均拥堵程度,对模型的拥堵计算区间进行优化和更新,实现高速路网拥堵识别的算法参数的自适应修正。该方法及系统可以提高高速公路拥堵识别的速度和准确性,并保证数据共享的安全性。


    Access

    Download


    Export, share and cite



    Title :

    Federal learning-based multi-modal highway congestion identification method and system


    Additional title:

    基于联邦学习的多模态高速公路拥堵识别方法及系统


    Contributors:
    YAO LINGYUN (author) / ZOU YING (author) / GUO JIANPING (author) / ZHENG YU (author) / TANG CHANGLIN (author)

    Publication date :

    2023-09-22


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G06V / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    Urban highway congestion

    Mcdonald, John F. | Online Contents | 1995



    Highway key information intelligent extraction system based on multi-modal deep learning

    LIU CONGYANG / LI HAIFENG / ZHANG HAIXIA et al. | European Patent Office | 2025

    Free access


    Highway congestion dissipation time prediction method based on deep meta-learning

    SUN DIHUA / ZHAO MIN / REN LING et al. | European Patent Office | 2023

    Free access