Atmospheric Turbulence Intensity Estimation Based on Deep Convolutional Neural Networks

Shengjie Ma; Shiqi Hao; Qingsong Zhao; Yong Wang; Lei Wang

doi:10.3788/CJL202148.0401018

Chinese Journal of Lasers, Volume. 48, Issue 4, 0401018(2021)

Atmospheric Turbulence Intensity Estimation Based on Deep Convolutional Neural Networks

Shengjie Ma^1,2, Shiqi Hao^1,2、*, Qingsong Zhao^1,2, Yong Wang^1,2, and Lei Wang^1,2

¹State Key Laboratory of Pulse Power Laser Technology, National University of Defense Technology, Hefei, Anhui 230037, China

²AnHui Province Key Laboratory of Electronic Restriction, Hefei,Anhui 230037, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(10)

Fig. 1. Model of multi-layer phase screen

Download full size

Fig. 2. Spot images of standard Gaussian beam and Gaussian beams under influence of atmospheric turbulence with different intensities. (a) Standard Gaussian beam; (b) $C_{n}^{2}$ =1.0083×10^-16 m^-2/3; (c) $<$

Download full size

Fig. 3. Structure of VGG16 model

Download full size

Fig. 4. Scatter plots, frequency distribution histograms, and cumulative probability distribution diagrams of estimation value and actual value of lg

C_{n}^{2}

by VGG16 model. (a1) (a2) (a3) Number of iterations is 1;(b1) (b2) (b3) number of iterations is 10;(c1) (c2) (c3) number of iterations is 20; (d

Download full size

Fig. 5. Scatter plots,frequency distribution histograms, and cumulative probability distribution diagrams of estimation value and actual value of lg $C_{n}^{2}$ by AlexNet model. (a1) (a2) (a3) Number of iterations is 1; (b1)(b2)(b3) number of iterations is 10; (c1)(c2)(c3) number of iterations is 20; (d1

Download full size

Table 1. Simulation parameters
View table
Table 1. Simulation parameters
Parameter Simulation value
λ /nm 632.8
w₀ /cm 2
z /m 1000
Δx/cm 0.04
Number of grid elements 768
Width of phase screen L /m 0.5
Distance of phase screen l /m 200
$C_{n}^{2}$ /m^-2/3 1.0×10^-16--1.0×10^-13

Table 2. Four statistics of VGG16 model under different numbers of iterations
View table
Table 2. Four statistics of VGG16 model under different numbers of iterations
Number of iterations E_MAE E_MRE E_RMSE R_xy
1 0.0845 0.1635 0.1099 92.32%
10 0.0432 0.0837 0.0577 98.16%
20 0.0272 0.0527 0.0349 99.24%
500 0.0133 0.0258 0.0171 99.84%

Table 3. Standard deviation of estimation results under different turbulence intensities
View table
Table 3. Standard deviation of estimation results under different turbulence intensities
$C_{n}^{2}$ 1.0×10^-16 m^-2/3 1.0×10^-15 m^-2/3 1.0×10^-14 m^-2/3 1.0×10^-13 m^-2/3
Standard deviation 0.071 0.0996 0.1453 0.2669

Table 4. Four statistics of AlexNet model under different numbers of iterations
View table
Table 4. Four statistics of AlexNet model under different numbers of iterations
Number of iterations E_MAE E_MRE E_RMSE R_xy
1 0.1902 0.3682 0.2325 59.38%
10 0.0866 0.1677 0.1149 94.80%
20 0.0691 0.1338 0.0856 97.41%
500 0.0269 0.0522 0.0340 99.27%

Table 5. Four statistics of two models under different numbers of iterations

View table

Table 5. Four statistics of two models under different numbers of iterations

Number of iterations	Model	E_MAE	E_MRE	E_RMSE	R_xy
1	AlexNet	0.1902	0.3682	0.2325	59.38%
1	VGG16	0.0845	0.1635	0.1099	92.32%
10	AlexNet	0.0866	0.1677	0.1149	94.80%
10	VGG16	0.0432	0.0837	0.0577	98.16%
20	AlexNet	0.0691	0.1338	0.0856	97.41%
20	VGG16	0.0272	0.0527	0.0349	99.24%
500	AlexNet	0.0269	0.0522	0.0340	99.27%
500	VGG16	0.0133	0.0258	0.0171	99.84%

Tools

Get Citation

Copy Citation Text

Shengjie Ma, Shiqi Hao, Qingsong Zhao, Yong Wang, Lei Wang. Atmospheric Turbulence Intensity Estimation Based on Deep Convolutional Neural Networks[J]. Chinese Journal of Lasers, 2021, 48(4): 0401018

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites