Chinese Journal of Liquid Crystals and Displays, Volume. 38, Issue 6, 770(2023)

Target event classification method through thinly scattered media

Bo-yu YANG and Jun KE*
Author Affiliations
  • School of Optics and Photonics, Beijing Institute of Technology,Beijing 100081,China
  • show less

    The existing target analysis through scattering media is mainly aimed at stationary targets. At the same time, ambient or external luminance conditions have a critical impact on the target analysis results. This paper puts forward extremely high requirements for the dynamic range and time resolution of the image acquisition equipment. Due to its high dynamic range, high temporal resolution and low latency, event cameras provide new solutions to the above problems. Aiming at the problem of poor detection effect of moving targets through thin scattering medium under low illumination, event camera is adopted to carry out target detection research through thin scattering medium. V2E network is used to prepare the “event speckle” dataset by grayscale speckle, and ResNet classification network is adopted for target classification. The ten-class accuracy of 94.27% is obtained. Experimental results show that the applicaiton of event streams has great potential for analyzing target information passing through scattering media.

    Tools

    Get Citation

    Copy Citation Text

    Bo-yu YANG, Jun KE. Target event classification method through thinly scattered media[J]. Chinese Journal of Liquid Crystals and Displays, 2023, 38(6): 770

    Download Citation

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

    Category: Research Articles

    Received: Jan. 31, 2023

    Accepted: --

    Published Online: Jun. 29, 2023

    The Author Email: Jun KE (jke@bit.edu.cn)

    DOI:10.37188/CJLCD.2023-0027

    Topics