• Infrared and Laser Engineering
  • Vol. 50, Issue 3, 20200344 (2021)
Hanyu Hong1、2、3, Shikang Wu1、2、3, Yu Shi1、2、3, Jinmeng Wu1、2、3, and Chunsheng Sun4
Author Affiliations
  • 1Hubei Key Laboratory of Optical Information and Pattern Recognition, Wuhan 430205, China
  • 2Hubei Research Centre of Video Image and High Denition Projection, Wuhan 430205, China
  • 3School of Electrical and Information Engineering, Wuhan Institute of Technology, Wuhan 430205, China
  • 4College of Weaponry Engineering, Naval University of Engineering, Wuhan 430032, China
  • show less
    DOI: 10.3788/IRLA20200344 Cite this Article
    Hanyu Hong, Shikang Wu, Yu Shi, Jinmeng Wu, Chunsheng Sun. Non-uniform strong noise removal method for non-cooperative mine target image[J]. Infrared and Laser Engineering, 2021, 50(3): 20200344 Copy Citation Text show less
    Flow chart of denoising algorithm
    Fig. 1. Flow chart of denoising algorithm
    Simulated mine target image
    Fig. 2. Simulated mine target image
    Strong noise statistics of simulated mine target image
    Fig. 3. Strong noise statistics of simulated mine target image
    Comparison of gradient and difference eigenvalue
    Fig. 4. Comparison of gradient and difference eigenvalue
    Gray distribution of edge perception constraint term 边缘感知约束项的灰度分布图
    Fig. 5. Gray distribution of edge perception constraint term 边缘感知约束项 的灰度分布图
    Comparison of denoising results of the two filtering algorithms
    Fig. 6. Comparison of denoising results of the two filtering algorithms
    PSNR comparison of the denoising results of the two filtering algorithms under different noise levels
    Fig. 7. PSNR comparison of the denoising results of the two filtering algorithms under different noise levels
    Comparison of denoising results of simulated mine target images
    Fig. 8. Comparison of denoising results of simulated mine target images
    Image segmentation comparison before and after denoising under different algorithms
    Fig. 9. Image segmentation comparison before and after denoising under different algorithms
    ImagesAlgorithmsNoise standard deviation
    152025303540
    Fig.6(a)LEP0.93780.92550.90740.88760.86150.8411
    LEP-EPC0.961480.94520.92050.90270.88910.8635
    Fig.6(d)LEP0.93190.92650.91310.89730.87560.8561
    LEP-EPC0.95940.94760.93250.91920.89210.8819
    Table 1. [in Chinese]
    AlgorithmsMine1Mine2Mine3Mine4
    ENLEPIENLEPIENLEPIENLEPI
    Original image33.671.039.101.031.651.08.331.0
    BM3D48.340.8448.360.5537.430.806.990.52
    L076.220.8273.750.6261.160.623.090.32
    LEP-EPCM422.750.98307.880.64230.750.86232.160.60
    Table 2. [in Chinese]
    AlgorithmsMine1Mine2Mine3Mine4
    BM3D12 s11.9 s11.7 s11.1 s
    L03.25 s3.27 s3.05 s3.07 s
    LEP-EPCM3.06 s3.08 s2.95 s3.01 s
    Table 3. [in Chinese]
    Hanyu Hong, Shikang Wu, Yu Shi, Jinmeng Wu, Chunsheng Sun. Non-uniform strong noise removal method for non-cooperative mine target image[J]. Infrared and Laser Engineering, 2021, 50(3): 20200344
    Download Citation