Traffic Velocity Estimation from Vehicle Count Sequences

Takayuki Katsuki, Tetsuro Morimura, Masato Inoue

    Research output: Contribution to journalArticle

    Abstract

    Traffic velocity is a fundamental metric for inferring traffic conditions. This paper proposes a new velocity estimation approach from temporal sequences of vehicle count that does not require tracking any vehicles or using any labeled data. It is useful for measuring traffic velocities with low quality and inexpensive sensors such as web cameras in general use. We formalize the task as a density estimation problem by introducing a new model for temporal sequences of vehicle counts wherein the correlation between the sequences is directly related to the traffic velocity. We also derive a sampling-based algorithm for the density estimation. We show the effectiveness of our method on artificial and real-world data sets.

    Original languageEnglish
    Article number7782816
    Pages (from-to)1700-1712
    Number of pages13
    JournalIEEE Transactions on Intelligent Transportation Systems
    Volume18
    Issue number7
    DOIs
    Publication statusPublished - 2017 Jul 1

    Fingerprint

    Cameras
    Sampling
    Sensors

    Keywords

    • Bayes procedures
    • Intelligent transportation systems
    • unsupervised learning
    • velocity measurement

    ASJC Scopus subject areas

    • Automotive Engineering
    • Mechanical Engineering
    • Computer Science Applications

    Cite this

    Traffic Velocity Estimation from Vehicle Count Sequences. / Katsuki, Takayuki; Morimura, Tetsuro; Inoue, Masato.

    In: IEEE Transactions on Intelligent Transportation Systems, Vol. 18, No. 7, 7782816, 01.07.2017, p. 1700-1712.

    Research output: Contribution to journalArticle

    Katsuki, Takayuki ; Morimura, Tetsuro ; Inoue, Masato. / Traffic Velocity Estimation from Vehicle Count Sequences. In: IEEE Transactions on Intelligent Transportation Systems. 2017 ; Vol. 18, No. 7. pp. 1700-1712.
    @article{814b0b1708f34ce2bf2a9a21daa576f4,
    title = "Traffic Velocity Estimation from Vehicle Count Sequences",
    abstract = "Traffic velocity is a fundamental metric for inferring traffic conditions. This paper proposes a new velocity estimation approach from temporal sequences of vehicle count that does not require tracking any vehicles or using any labeled data. It is useful for measuring traffic velocities with low quality and inexpensive sensors such as web cameras in general use. We formalize the task as a density estimation problem by introducing a new model for temporal sequences of vehicle counts wherein the correlation between the sequences is directly related to the traffic velocity. We also derive a sampling-based algorithm for the density estimation. We show the effectiveness of our method on artificial and real-world data sets.",
    keywords = "Bayes procedures, Intelligent transportation systems, unsupervised learning, velocity measurement",
    author = "Takayuki Katsuki and Tetsuro Morimura and Masato Inoue",
    year = "2017",
    month = "7",
    day = "1",
    doi = "10.1109/TITS.2016.2628384",
    language = "English",
    volume = "18",
    pages = "1700--1712",
    journal = "IEEE Transactions on Intelligent Transportation Systems",
    issn = "1524-9050",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",
    number = "7",

    }

    TY - JOUR

    T1 - Traffic Velocity Estimation from Vehicle Count Sequences

    AU - Katsuki, Takayuki

    AU - Morimura, Tetsuro

    AU - Inoue, Masato

    PY - 2017/7/1

    Y1 - 2017/7/1

    N2 - Traffic velocity is a fundamental metric for inferring traffic conditions. This paper proposes a new velocity estimation approach from temporal sequences of vehicle count that does not require tracking any vehicles or using any labeled data. It is useful for measuring traffic velocities with low quality and inexpensive sensors such as web cameras in general use. We formalize the task as a density estimation problem by introducing a new model for temporal sequences of vehicle counts wherein the correlation between the sequences is directly related to the traffic velocity. We also derive a sampling-based algorithm for the density estimation. We show the effectiveness of our method on artificial and real-world data sets.

    AB - Traffic velocity is a fundamental metric for inferring traffic conditions. This paper proposes a new velocity estimation approach from temporal sequences of vehicle count that does not require tracking any vehicles or using any labeled data. It is useful for measuring traffic velocities with low quality and inexpensive sensors such as web cameras in general use. We formalize the task as a density estimation problem by introducing a new model for temporal sequences of vehicle counts wherein the correlation between the sequences is directly related to the traffic velocity. We also derive a sampling-based algorithm for the density estimation. We show the effectiveness of our method on artificial and real-world data sets.

    KW - Bayes procedures

    KW - Intelligent transportation systems

    KW - unsupervised learning

    KW - velocity measurement

    UR - http://www.scopus.com/inward/record.url?scp=85006867179&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85006867179&partnerID=8YFLogxK

    U2 - 10.1109/TITS.2016.2628384

    DO - 10.1109/TITS.2016.2628384

    M3 - Article

    VL - 18

    SP - 1700

    EP - 1712

    JO - IEEE Transactions on Intelligent Transportation Systems

    JF - IEEE Transactions on Intelligent Transportation Systems

    SN - 1524-9050

    IS - 7

    M1 - 7782816

    ER -