Infrared and Laser Engineering, Volume. 53, Issue 5, 20240049(2024)

Deep learning-based infrared imaging degradation model identification and super-resolution reconstruction

Junfeng Cao1,2,3,4, Qinghai Ding5, Depeng Zou6, Hengjia Qin6, and Haibo Luo1,2,3
Author Affiliations
  • 1Key Laboratory of Opto-Electronic Information Processing, Chinese Academy of Sciences, Shenyang 110016, China
  • 2Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China
  • 3Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110169, China
  • 4University of Chinese Academy of Sciences, Beijing 100049, China
  • 5Space Star Technology Co., Ltd., Beijing 100086, China
  • 6The Third Military Representative Office of the Air Force Equipment Department in Shenyang, Shenyang 110016, China
  • show less
    Figures & Tables(12)
    Target image acquisition system
    Calibration target
    Architecture of blur kernel modeling network
    Architecture of super-resolution network
    Target images acquired at different working temperatures
    Blur kernel varies with working temperature and spatial position
    (a) Calibrated blur kernels; (b) Blur kernels predicted by blur kernel modeling network
    Prediction accuracy of blur kernel modeling network
    (a) Real-world image acquisition system; (b) Experimental set up inside the test chamber
    Visual results of different methods on “scene 1” images captured at different working temperature for scale factor 4
    Visual results of different methods on “scene2” images captured at different working temperature for scale factor 4
    • Table 1. Quantitative comparison with other methods with scale factor 4

      View table
      View in Article

      Table 1. Quantitative comparison with other methods with scale factor 4

      MethodNIQE↓PIQE↓BRISQUE↓
      Bicubic6.09192.84452.300
      IKC5.00585.70651.981
      MANet+RRDB-SFT5.53487.48852.561
      KernelGAN+USRNet4.27079.22250.164
      Ours4.24082.09049.614
    Tools

    Get Citation

    Copy Citation Text

    Junfeng Cao, Qinghai Ding, Depeng Zou, Hengjia Qin, Haibo Luo. Deep learning-based infrared imaging degradation model identification and super-resolution reconstruction[J]. Infrared and Laser Engineering, 2024, 53(5): 20240049

    Download Citation

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

    Category:

    Received: Jan. 27, 2024

    Accepted: --

    Published Online: Jun. 21, 2024

    The Author Email:

    DOI:10.3788/IRLA20240049

    Topics