Laser & Optoelectronics Progress, Volume. 58, Issue 20, 2015006(2021)

Anomaly Event Detection Based on Two-Stream Network and Multi-instance Learning

Xianbin Yang1, Jianwu Dang1,2、*, Song Wang1,2, and Yangping Wang2,3
Author Affiliations
  • 1School of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou, Gansu 730070, China
  • 2Gansu Provincial Engineering Research Center for Artificial Intelligence and Graphic & Image Processing, Lanzhou, Gansu 730070, China;
  • 3National Experimental Teaching Demonstration Center of Computer Science and technology, Lanzhou Jiaotong University, Lanzhou, Gansu 730070, China
  • show less
    Figures & Tables(7)
    Flow chart of the proposed anomaly detection method
    ROCs of different anomaly detection methods
    Comparison charts of qualitative results for video frame tests of various anomaly events. (a) Explosion; (b) animal abuse; (c) traffic accident; (d) normal video frame fragment
    Score variations of anomaly video clips under different iteration times. (a) 2000 times; (b) 5000 times; (c) 8000 times; (d) 12000 times
    • Table 1. Configuration of experimental environment

      View table

      Table 1. Configuration of experimental environment

      Experimental environmentDetailed information
      CPU: Intel(R) Core(TM) i7-8700 at 3.20 GHz
      ComputerRAM: 16 GB
      MATLAB 2017
      Python 2.7
      GPU: NVIDIA GeForce GTX 1060 3GB
      ServerTensorFlow 1.6
      Display memory: 12 GB
      Ubuntu 14.04
    • Table 2. Comparison of experimental results of two models

      View table

      Table 2. Comparison of experimental results of two models

      ModelNetworkAUCDetection speed / fps
      Time flow: RGB71.43
      I3DSpatial flow: TV-L173.61
      Two-stream fusion model: TV-L1+RGB76.8325
      Time flow: RGB71.43
      M-I3DSpatial flow: MotionNet74.84
      Two-stream fusion model: MotionNet +RGB77.2532
    • Table 3. Comparisons of the experimental results between the proposed method and other methods

      View table

      Table 3. Comparisons of the experimental results between the proposed method and other methods

      MethodEER /%AUCDetection speed /fps
      Method in Ref. [23]34.950.66
      Method in Ref. [5]27.365.51143.50
      Method in Ref. [25]23.669.300.04
      Method in Ref. [21]75.41
      Method in Ref. [24]18.173.68120.00
      Our method16.877.2532.00
    Tools

    Get Citation

    Copy Citation Text

    Xianbin Yang, Jianwu Dang, Song Wang, Yangping Wang. Anomaly Event Detection Based on Two-Stream Network and Multi-instance Learning[J]. Laser & Optoelectronics Progress, 2021, 58(20): 2015006

    Download Citation

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

    Category: Machine Vision

    Received: Nov. 25, 2020

    Accepted: Jan. 20, 2021

    Published Online: Oct. 14, 2021

    The Author Email: Dang Jianwu (dangjw@mail.lzjt.cn)

    DOI:10.3788/LOP202158.2015006

    Topics