Journal of Infrared and Millimeter Waves, Volume. 42, Issue 4, 546(2023)

The infrared point target detection algorithm based on modified random walker and non-convex rank approximation minimization under the complex background

Kun WANG1,2, De-Fu JIANG2、*, Li-Jun YUN1, and Ling-Fan WU3
Author Affiliations
  • 1School of Information Science and Technology,Yunnan Normal University,Kunming 650500,China
  • 2College of Computer and Information,Hohai University,Nanjing 211100,China
  • 3Institute of Microelectronic Technology of Kunshan,Chinese Academy of Sciences,Kunshan215347,China
  • show less
    Figures & Tables(11)
    (a)The infrared image with a single and smooth background,(b)the infrared image with a complex background,(c)the normalized gray level image of the confidence mapping of(a)the corresponding infrared image,(d)the normalized gray level image of the confidence mapping of(b)the corresponding infrared image,(e)the three-dimensional grid image of(c),(f)the three-dimensional grid image of(d)
    (a)and(b)are the original images containing infrared point targets respectively,(c)and(d)represent the normalized gray level images of the confidence mapping of the infrared images corresponding to(a)and(b),(e)and(f)are the three-dimensional grid images of the confidence mapping
    The flow of the NRAM-MRW algorithm
    The local region of a small target in an infrared image
    The infrared images and the detection results by the LCM and MPCM algorithms
    The infrared images and the 3D grid images of the detection results by LCM and MPCM algorithms
    The infrared images and the detection results of 5 algorithms
    The infrared images and the 3D grid images of the detection results by the 5 algorithms
    (a)The ROC curve of data6 sequence,(b)the ROC curve of data10 sequence,(c)the ROC curve of data13 sequence,(d)the ROC curve of data14 sequence,(e)the ROC curve of data17 sequence
    • Table 1. The detailed descriptions of 5 images for the selected dates

      View table
      View in Article

      Table 1. The detailed descriptions of 5 images for the selected dates

      图号数据段帧号目标位置坐标场景描述
      adata6235(102,94)地空交界背景
      bdata10215(111,88)地空交界背景
      cdata13452(101,102)含有目标微弱
      ddata141234(83,50)有地面车辆干扰
      edata1736(92,48)含有目标微弱
    • Table 2. The BSF and LCG values of each algorithm of 5 infrared images

      View table
      View in Article

      Table 2. The BSF and LCG values of each algorithm of 5 infrared images

      Methodsabcde
      BSFLCGBSFLCGBSFLCGBSFLCGBSFLCG
      PSTNN9.6620.867InfInfInfInfInfInfInfInf
      IPI10.0531.0768.0530.916InfInfInfInf5.2150.326
      FKRWInfInfInfInfInfInfInfInfInfInf
      NRAM16.4461.63516.2281.28618.6431.23118.1061.26417.3381.186
      NRAM-MRW22.8621.57620.5641.51323.2861.83522.4691.76921.2531.634
    Tools

    Get Citation

    Copy Citation Text

    Kun WANG, De-Fu JIANG, Li-Jun YUN, Ling-Fan WU. The infrared point target detection algorithm based on modified random walker and non-convex rank approximation minimization under the complex background[J]. Journal of Infrared and Millimeter Waves, 2023, 42(4): 546

    Download Citation

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

    Category: Research Articles

    Received: Sep. 12, 2022

    Accepted: --

    Published Online: Aug. 1, 2023

    The Author Email: De-Fu JIANG (Surfer_jiangdf0801@163.com)

    DOI:10.11972/j.issn.1001-9014.2023.04.017

    Topics