A graphical method for detection of Flash Crowds in traffic

Marat Zhanikeev, Yoshiaki Tanaka

    Research output: Contribution to journalArticle

    1 Citation (Scopus)

    Abstract

    Majority of traffic analysis tools today are based on NetFlow technology including its more recent successor, IPFIX protocol. Even with the flexibility of IPFIX in mind, processing traffic in realtime is still difficult. Given the urgent need for lightweight methods in the area, this paper is crossing inter-disciplinary borders to find a solution. Specifically, this paper looks into a possibility of applying video compression to 2D visualization of traffic in search for anomalies. The proposed method is applied to detection of Flash Crowds in traffic and is found successful when compared to other methods.

    Original languageEnglish
    Pages (from-to)91-105
    Number of pages15
    JournalTelecommunication Systems
    Volume57
    Issue number1
    DOIs
    Publication statusPublished - 2014

      Fingerprint

    Keywords

    • Anomaly detection
    • Flash Crowd
    • Traffic analysis
    • Video compression

    ASJC Scopus subject areas

    • Electrical and Electronic Engineering

    Cite this