Laser & Optoelectronics Progress, Volume. 57, Issue 6, 061506(2020)

Detection Algorithm of Crowd Abnormal Event Based on Girvan-Newman Splitting

Wentao Li, han Fu, Zhen Hao, Yan Ten, Lin Yan, Peiran Zhao, and Xuewu Zhang*
Author Affiliations
  • College of Internet of Things Engineering, Hohai University, Changzhou, Jiangsu 213022, China
  • show less
    Figures & Tables(8)
    Flow chart of abnormal event detection algorithm based on group analysis
    Group division diagram. (a) Connection establishment; (b) group division
    Grouping result. (a) Scene 1; (b) scene 2
    Experimental scene 1. (a) Normal scene; (b) abnormal scene
    Experimental scene 2. (a) Normal scene; (b) abnormal scene
    Experimental scene 3. (a) Normal scene; (b) abnormal scene
    • Table 1. Abnormal event detection experiment resultuint: frame

      View table

      Table 1. Abnormal event detection experiment resultuint: frame

      AlgorithmTruevalueDetection
      abnormlnormal
      HOFabnorml721117
      normal1506325
      DTabnorml73999
      normal906385
      EMabnorml635203
      normal2106265
      Ouralgorithmabnorml76672
      normal1216354
    • Table 2. Performance comparison of abnormal event detection algorithmsunit: %

      View table

      Table 2. Performance comparison of abnormal event detection algorithmsunit: %

      AlgorithmAccuracyAccuracyrateRecallrateSpecificityF1score
      HOF96.382.886.097.784.4
      DT97.489.188.298.688.7
      EM94.475.175.896.875.5
      Our97.486.491.498.188.8
    Tools

    Get Citation

    Copy Citation Text

    Wentao Li, han Fu, Zhen Hao, Yan Ten, Lin Yan, Peiran Zhao, Xuewu Zhang. Detection Algorithm of Crowd Abnormal Event Based on Girvan-Newman Splitting[J]. Laser & Optoelectronics Progress, 2020, 57(6): 061506

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Machine Vision

    Received: Aug. 20, 2019

    Accepted: Sep. 6, 2019

    Published Online: Mar. 6, 2020

    The Author Email: Zhang Xuewu (Lab_112@126.com)

    DOI:10.3788/LOP57.061506

    Topics