• Acta Photonica Sinica
  • Vol. 50, Issue 12, 1210001 (2021)
Yifan HAO1、2、3 and Yi JIAN1、2、*
Author Affiliations
  • 1Key Laboratory of Infrared Detection and Imaging Technology, Chinese Academy of; Sciences, Shanghai 200083, China
  • 2Shanghai Institute of Technical Physics, Chinese Academy of Sciences, Shanghai 200083, China
  • 3University of Chinese Academy of Sciences, Beijing 100049, China
  • show less
    DOI: 10.3788/gzxb20215012.1210001 Cite this Article
    Yifan HAO, Yi JIAN. Multi-focus Fusion Method Based on Multi-Scale Local Weighted Variance[J]. Acta Photonica Sinica, 2021, 50(12): 1210001 Copy Citation Text show less
    Flowchart of multi-focus fusion
    Fig. 1. Flowchart of multi-focus fusion
    The output of the intermediate result of the original image quantization definition (there are 30 original images corresponding to the data)
    Fig. 2. The output of the intermediate result of the original image quantization definition (there are 30 original images corresponding to the data)
    The influence of mask morphology corrosion factor size to the final result
    Fig. 3. The influence of mask morphology corrosion factor size to the final result
    Intermediate result of the refined mask
    Fig. 4. Intermediate result of the refined mask
    The result R of multi-focus fusion
    Fig. 5. The result R of multi-focus fusion
    Part of the original image
    Fig. 6. Part of the original image
    Multi focus fusion results of three experiments
    Fig. 7. Multi focus fusion results of three experiments
    Partial enlarged view of Fig. 7(a)~(d)
    Fig. 8. Partial enlarged view of Fig. 7(a)~(d)
    Quality score line chart
    Fig. 9. Quality score line chart
    ENMA′s methodZHU′s methodQIU′s methodProposed method
    Experiment 17.304 27.318 77.473 57.480 4
    Experiment 27.203 36.614 27.317 06.888 3
    Experiment 36.940 86.829 56.708 85.845 1
    Table 1. Information entropy(EN) of the fusion results of each algorithm
    PSNRMA′s methodZHU′s methodQIU′s methodProposed method
    Experiment 124.111 924.096 025.247 625.312 3
    Experiment 216.190 118.977 913.832 617.550 5
    Experiment 318.917 319.703 417.885 522.534 0
    Table 2. Peak signal-to-noise ratio(PSNR) of the fusion results of each algorithm
    SSIMMA′s methodZHU′s methodQIU′s methodProposed method
    Experiment 129.984 329.984 429.988 329.988 9
    Experiment 2104.650 0104.804 6104.389 7104.735 1
    Experiment 319.973 019.974 819.956 819.986 1
    Table 3. Structural similarity index measure(SSIM) of the fusion results of each algorithm
    CEMA′s methodZHU′s methodQIU′s methodProposed method
    Experiment 14.753 25.747 91.350 81.317 8
    Experiment 2293.836 8253.182 9300.672 4235.142 7
    Experiment 313.219 18.782 528.137 88.780 8
    Table 4. Cross entropy(CE) of the fusion results of each algorithm
    PSMA′s methodZHU′s methodQIU′s methodProposed method
    Experiment 10.056 10.058 70.064 50.066 7
    Experiment 20.071 80.040 10.090 60.058 1
    Experiment 30.058 80.057 00.052 80.060 1
    Table 5. Perceptual saliency (PS) of the fusion results of each algorithm
    Yifan HAO, Yi JIAN. Multi-focus Fusion Method Based on Multi-Scale Local Weighted Variance[J]. Acta Photonica Sinica, 2021, 50(12): 1210001
    Download Citation