To avoid the difficulty of collecting accurate traffic flow data, this paper proposes a novel approach for congestion features extraction from traffic video. The approach firstly segments the traffic video into shots and the shot motion content feature is extracted. Then, we extract the key frames applying an improved global k-means clustering algorithm. The last congestion feature of the global optical flow energy is computed based on the key frames. The numerical experiments on traffic surveillance video show the validity and high accuracy for traffic congestion detection using the propose method in this paper


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    A Feature Extraction Approach of Traffic Congestion from Video



    Published in:

    Advanced Materials Research ; 490-495 ; 1058-1062


    Publication date :

    2012-03-15


    Size :

    5 pages




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English





    An Automatic Recognition Approach for Traffic Congestion States Based on Traffic Video

    Zou, Fu-min / Llao, Lü-chao / Jiang, Xin-hua et al. | ASCE | 2014



    Abnormal Traffic Congestion Recognition Based on Video Analysis

    Liu, Xueli / Gao, Wen / Feng, Dong et al. | IEEE | 2020


    Traffic Lane Congestion Ratio Evaluation by Video Data

    Stetsenko, Inna V. / Stelmakh, Oleksandr | Springer Verlag | 2019