Nonlinear Reconstruction for Target Density Based on Randomly Perturbed Optimization and Multi-models Fusion

Jinxin XU; Qingwu LI; Zhiqiang GUAN; Xiaolin WANG

doi:10.3788/gzxb20225103.0310006

Acta Photonica Sinica, Volume. 51, Issue 3, 0310006(2022)

Nonlinear Reconstruction for Target Density Based on Randomly Perturbed Optimization and Multi-models Fusion

Jinxin XU1...2, Qingwu LI2,*, Zhiqiang GUAN1 and Xiaolin WANG2 |Show fewer author(s)

Author Affiliations

¹Nanjing Marine Radar Institute，Nanjing 211106，China

²College of Internet of Things Engineering，Hohai University，Changzhou ，Jiangsu 213002，China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(12)

Fig. 1. First derivative products of different forward models

Download full size

View in Article

Fig. 2. ACF and IACT of sample estimation for each algorithm

Download full size

View in Article

Fig. 3. The RE of sample estimation and reconstructed time by each algorithm

Download full size

View in Article

Fig. 4. The RE involving multi-model fusion

Download full size

View in Article

Fig. 5. Reconstructed results of FTO transmission image with variable density （1.25% noise level）

Download full size

View in Article

Fig. 6. Reconstructed results of FTO transmission image with constant density （1.25% noise level）

Download full size

View in Article

Fig. 7. High energy flash X-ray static image under 4 MeV

Download full size

View in Article

Fig. 8. Results of density reconstruction （4 MeV static image）

Download full size

View in Article

Fig. 9. Comparison of reconstructed density profiles of each algorithm

Download full size

View in Article

Table 1. The PSNR and SSIM of reconstructed results of constant density FTO images

View table

View in Article

Table 1. The PSNR and SSIM of reconstructed results of constant density FTO images

SPA

LRIS_

Gamma

LRIS_

Jeffrey

TLE_Gibbs

S_RTO

Proposed

（T_F）

Proposed

（I_nF）

Proposed

method

18.074 4

/0.665 4

19.479 2

/0.825 4

19.448

/0.826 2

20.199 5

/0.713 9

24.504 1

/0.897 0

53.267 0

/0.999 7

19.099 8

/0.899 6

25.141 0

/0.924 7

0.25%

17.913 7

/0.619 8

19.489 6

/0.792 8

19.461 2

/0.794 9

20.284 4

/0.695 8

23.414 6

/ 0.886 1

28.116 1

/0.882 6

19.096 9

/0.888 4

25.150 9

/0.918 2

0.75%

16.836

/0.528

19.573 6

/0.705 8

19.554 2

/0.704 1

20.719 5

/0.644 4

23.788 1

/0.802 2

25.632 1

/0.827 2

19.030 1

/0.821 6

25.181 7

/0.880 3

1.25%

14.583

/0.457 9

19.869 8

/0.661 6

19.850 5

/0.660 4

20.864 7

/0.603 7

22.913 6

/0.738 3

24.590 6

/0.744 6

18.870 6

/0.750 5

24.743 3

/0.840 2

Table 2. The PSNR and SSIM of reconstructed results of variable density FTO images

View table

View in Article

Table 2. The PSNR and SSIM of reconstructed results of variable density FTO images

SPA

LRIS_

Gamma

LRIS_

Jeffrey

TLE_Gibbs

S_RTO

Proposed

（T_F）

Proposed

（I_nF）

Proposed

method

19.126 7

/0.826 5

19.865 2

/0.911 1

19.833 2

/0.910 9

21.047 1

/0.928 9

26.663 7

/0.939 6

54.532 1

/0.999 7

19.661 9

/0.950 2

27.650 7

/0.960 0

0.25%

19.037 2

/0.787 8

19.896 5

/0.895

19.845

/0.894 6

21.066 1

/0.799 1

26.714 4

/0.927 4

28.890 6

/0.933 7

19.666 7

/0.939 5

27.578 5

/0.953 8

0.75%

18.306 9

/0.615 2

19.892 9

/0.811 9

19.835 7

/0.807 6

20.923 9

/0.719 3

25.816 5

/0.882 4

27.499 7

/0.897 0

19.587 8

/0.872 8

27.883 8

/0.934 6

1.25%

17.001 4

/0.469 9

20.048

/0.744 1

20.012 8

/0.741 6

21.449 4

/0.676 6

25.270 6

/0.826 1

26.624 8

0.864 0

19.567 5

/0.784 4

27.579 1

/0.903 6

Table 3. Average value （g/cm³）and relative error （%）of density reconstruction of each algorithm

View table

View in Article

Table 3. Average value （g/cm³）and relative error （%）of density reconstruction of each algorithm

	GPSR	LRIS_Jeffrey	LRIS_Gamma	SPA	TLE_Gibbs	Proposed method
CO1	7.592 6/5.692 1	7.593 0/5.032 6	7.594 7/5.065 6	7.571 3/5.327 7	7.567 4/5.268 1	7.391 7/2.520 7
CO2	7.335 3/4.737 4	7.398 4/2.664 4	7.396 4/2.661 2	7.367 0/2.540 5	7.360 5/2.431 1	7.308 6/2.376 5

Tools

Get Citation

Copy Citation Text

Jinxin XU, Qingwu LI, Zhiqiang GUAN, Xiaolin WANG. Nonlinear Reconstruction for Target Density Based on Randomly Perturbed Optimization and Multi-models Fusion[J]. Acta Photonica Sinica, 2022, 51(3): 0310006

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category:

Received: Aug. 11, 2021

Accepted: Sep. 14, 2021

Published Online: Apr. 8, 2022

The Author Email: LI Qingwu (li_qingwu@163.com)

DOI:10.3788/gzxb20225103.0310006

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology

Table 1. The PSNR and SSIM of reconstructed results of constant density FTO images

Table 1. The PSNR and SSIM of reconstructed results of constant density FTO images

Table 2. The PSNR and SSIM of reconstructed results of variable density FTO images

Table 2. The PSNR and SSIM of reconstructed results of variable density FTO images

Table 3. Average value （g/cm3）and relative error （%）of density reconstruction of each algorithm

Table 3. Average value （g/cm3）and relative error （%）of density reconstruction of each algorithm

微信扫一扫：分享

Table 3. Average value （g/cm³）and relative error （%）of density reconstruction of each algorithm

Table 3. Average value （g/cm³）and relative error （%）of density reconstruction of each algorithm