Acta Photonica Sinica, Volume. 53, Issue 11, 1110002(2024)

Noise Image Processing Algorithm for a Low-light EBCMOS Acquisition System Based on FPGA

Yi CAO1,2, Xiangping ZHU1、*, Xiaomo ZHANG1,2, Wei ZHAO1, and Jun MA3
Author Affiliations
  • 1State Key Laboratory of Transient Optics and Photonics,Xi'an Institute of Optics and Precision Mechanics,Chinese Academy of Sciences,Xi'an 710119,China
  • 2University of Chinese Academy of Sciences,Beijing 100049,China
  • 3Xi'an Institute of Atomic Precision Manufacturing Technology Co.,Ltd.,Xi'an 710110,China
  • show less
    Figures & Tables(21)
    System block diagram
    Algorithm flowchart
    3×3 template generation
    3×3 template generation circuit diagram
    Median filter sorting
    Timing diagram for sorting simulation
    Gaussian filter template
    Sobel operator template
    Laplacian sharpening template
    Simulation test image
    Noise reduction comparison image at 10-1 lx
    Noise reduction comparison image at 10-2 lx
    Noise reduction comparison image at 10-4 lx
    Hardware development platform
    Actual shot of EBCMOS
    Comparison of processing time for a single frame image between software and hardware platforms
    Resource consumption of the algorithm on FPGA in this study
    • Table 1. Peak signal-to-noise ratio of the denoised image at 10-1 lx test

      View table
      View in Article

      Table 1. Peak signal-to-noise ratio of the denoised image at 10-1 lx test

      Filtering algorithmsNoise mapMedian filteringGaussian filteringRef. [17Ref. [22Proposed algorithm
      Lena19.421 331.96628.93028.82433.00636.834
      Peppers19.050 931.13128.33027.56232.23736.038
      Goldhill21.664 931.13128.33027.398731.01634.222
    • Table 2. Peak signal-to-noise ratio of the denoised image at 10-2 lx test

      View table
      View in Article

      Table 2. Peak signal-to-noise ratio of the denoised image at 10-2 lx test

      Filtering algorithmsNoise mapMedian filteringGaussian filteringRef. [17Ref. [22Proposed algorithm
      Lena19.34728.55827.19528.66928.73735.714
      Peppers18.88333.88029.75930.90634.79135.293
      Goldhill19.10733.14529.41631.26734.41735.125
    • Table 3. Peak signal-to-noise ratio of the denoised image at 10-4 lx test

      View table
      View in Article

      Table 3. Peak signal-to-noise ratio of the denoised image at 10-4 lx test

      Filtering algorithmsNoise mapMedian filteringGaussian filteringRef. [17Ref. [22Proposed algorithm
      Lena18.54135.65228.16534.32136.77639.478
      Peppers18.26634.76828.63433.65736.52737.545
      Goldhill18.40434.13229.62832.97235.399437.186
    • Table 4. Comparison of software and hardware platforms

      View table
      View in Article

      Table 4. Comparison of software and hardware platforms

      PSNRSoftware platformFPGAError
      1×10-1 lx noise map30.25730.1250.436%
      1×10-2 lx noise map33.78733.5040.517%
      1×10-4 lx noise map38.61138.3950.563%
    Tools

    Get Citation

    Copy Citation Text

    Yi CAO, Xiangping ZHU, Xiaomo ZHANG, Wei ZHAO, Jun MA. Noise Image Processing Algorithm for a Low-light EBCMOS Acquisition System Based on FPGA[J]. Acta Photonica Sinica, 2024, 53(11): 1110002

    Download Citation

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

    Category:

    Received: Mar. 1, 2024

    Accepted: Apr. 24, 2024

    Published Online: Jan. 8, 2025

    The Author Email: Xiangping ZHU (xpzhu@opt.ac.cn)

    DOI:10.3788/gzxb20245311.1110002

    Topics