Laser & Optoelectronics Progress, Volume. 61, Issue 22, 2237008(2024)

Infrared Tiny Target Detection Method Based on a Multi-Hop Deep Network

Shule Yan1, Runyu Chen1, Nian Cai1、*, Shaoqiu Xu1, and Jian Chen2
Author Affiliations
  • 1School of Information Engineering, Guangdong University of Technology, Guangzhou 510006, Guangdong , China
  • 2Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, Jilin , China
  • show less
    Figures & Tables(9)
    Imaging characteristics of infrared targets in complex background. (a) Target imaging performance; (b) grayscale distribution of infrared images; (c) top view of grayscale distribution
    Overall framework of proposed method
    Structure of MMFB and class activation mapping diagram of corresponding branches
    ROC curves for different compared methods
    Visualization of detection results
    Visualization of infrared targets on IRDST dataset
    • Table 1. Comparisons of different methods on infrared target detection

      View table

      Table 1. Comparisons of different methods on infrared target detection

      MethodTargetRecall /%Precision /%F1-score /%Fa /%FNR /%Time /ms
      MDFA-CGANTiny56.852.454.541.643.227.1
      Small71.145.455.416.528.9
      ACMTiny19.926.722.844.180.130.3
      Small54.957.456.17.845.1
      DNANetTiny69.973.671.720.230.131.3
      Small86.886.786.82.613.2
      YOLO-FRTiny92.160.673.148.37.96.4
      Small90.656.369.513.59.4
      Proposed methodTiny97.094.595.74.53.011.3
      Small91.196.493.70.78.9
    • Table 2. Ablation experiments

      View table

      Table 2. Ablation experiments

      BaselineMMFBNWDTargetRecallPrecisionF1-scoreFaFNR
      Tiny95.179.086.320.44.9
      Small46.599.663.419.553.5
      Tiny95.984.790.04.24.1
      Small89.489.789.62.010.6
      Tiny92.090.491.25.38.0
      Small93.582.687.73.86.5
      Tiny97.094.595.74.53.0
      Small91.196.493.70.78.9
    • Table 3. Comparison on IRDST dataset

      View table

      Table 3. Comparison on IRDST dataset

      MethodRecallPrecisionF1-scoreFaFNR
      MDFA-CGAN41.644.843.155.258.4
      ACM35.535.935.764.164.5
      DNANet89.290.589.99.510.8
      YOLO-FR83.258.568.741.516.8
      Proposed method93.695.794.64.36.4
    Tools

    Get Citation

    Copy Citation Text

    Shule Yan, Runyu Chen, Nian Cai, Shaoqiu Xu, Jian Chen. Infrared Tiny Target Detection Method Based on a Multi-Hop Deep Network[J]. Laser & Optoelectronics Progress, 2024, 61(22): 2237008

    Download Citation

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

    Category: Digital Image Processing

    Received: Jan. 25, 2024

    Accepted: Apr. 3, 2024

    Published Online: Nov. 19, 2024

    The Author Email: Nian Cai (cainian@gdut.edu.cn)

    DOI:10.3788/LOP240610

    CSTR:32186.14.LOP240610

    Topics