Optoelectronic Technology, Volume. 43, Issue 3, 276(2023)

A Violation Behavior Detection Algorithm Based on Yolov5 and Dilb for Online Video Surveillance

Yan FENG1 and Lei LU2
Author Affiliations
  • 1Xi 'an Polytechnic University, Xi'an 70048, CHN
  • 2School of Information and Communication Engineering, Xi'an Jiaotong University, Xi'an 710049, CHN
  • show less
    Figures & Tables(11)
    Algorithm flow diagram
    Improved ShuffleNetV2 with two blocks
    Flow diagram of proxy test behavior detection
    Schematic of lip area modeling
    Schematic of head pose modeling
    Annotation process and partial images
    Examination results for unrelated personnel and items
    Test results of candidate substitution
    Test results of candidates' speaking detection
    Test results of examinees' attention drift detection
    • Table 1. Performance comparison of algorithms

      View table
      View in Article

      Table 1. Performance comparison of algorithms

      方法mAP@0.5FPS(帧/s)
      YOLOv464.122.8
      YOLOv566.337.6
      Faster R‑CNN67.213.1
      SSD55.715.3
      FS‑YOLOv5(ours)69.636.3
    Tools

    Get Citation

    Copy Citation Text

    Yan FENG, Lei LU. A Violation Behavior Detection Algorithm Based on Yolov5 and Dilb for Online Video Surveillance[J]. Optoelectronic Technology, 2023, 43(3): 276

    Download Citation

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

    Category:

    Received: Jul. 15, 2023

    Accepted: --

    Published Online: Mar. 21, 2024

    The Author Email:

    DOI:10.19453/j.cnki.1005-488x.2023.03.015

    Topics