Underwater Turbulence Detection Technology Based on Convolutional Neural Networks

Fengtao HE; Qianqian WU; Jianlei ZHANG; Yi YANG; Juan ZHANG; Xinyu YAO; Weilin ZHAO

doi:10.3788/gzxb20245301.0111004

Acta Photonica Sinica, Volume. 53, Issue 1, 0111004(2024)

Underwater Turbulence Detection Technology Based on Convolutional Neural Networks

Fengtao HE^*, Qianqian WU, Jianlei ZHANG, Yi YANG, Juan ZHANG, Xinyu YAO, and Weilin ZHAO

Author Affiliations

School of Electronic Engineering，Xi'an University of Posts and Telecommunications，Xi'an 710072，China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(15)

Fig. 1. Standard speckle images and speckle images under the influence of underwater turbulence of different intensities

Download full size

View in Article

Fig. 2. Light intensity distribution

Download full size

View in Article

Fig. 3. Overall architecture

Download full size

View in Article

Fig. 4. Feature image

Download full size

View in Article

Fig. 5. Deep separable ResNet model

Download full size

View in Article

Fig. 6. Transformer mechanism

Download full size

View in Article

Fig. 7. The classification accuracy curve and loss value curve of the simulated dataset on this model

Download full size

View in Article

Fig. 8. The performance of different classification methods in terms of accuracy，precision，and recall

Download full size

View in Article

Fig. 9. Experimental device

Download full size

View in Article

Fig. 10. The classification accuracy curve and loss value curve of the experimental dataset on this model

Download full size

View in Article

Fig. 11. The classification accuracy curve and loss value curve of the experimental dataset on this model

Download full size

View in Article

Fig. 12. The performance of the experimental dataset in terms of accuracy and recall on this model

Download full size

View in Article

Table 1. Simulation parameters
View table
View in Article
Table 1. Simulation parameters
Parameter Simulation value
$λ / n m$ 523
$η / m$ 0.01
$w$ -5
$z / m$ 1 000
$W i d t h o f p h a s e s c r e e n L / m$ 0.5
$D i s t a n c e o f p h a s e s c r e e n l / m$ 200
$X_{T} / K^{2} s^{- 1}$ 10^-4~10^-8

Table 2. $X_{T}$ performance indicators of different classification networks in simulated datasets

View table

View in Article

Table 2. $X_{T}$ performance indicators of different classification networks in simulated datasets

Model	Temperature/℃	Precision/%	Recall/%	Accuracy/%
ResNet	10^-4	98.0	98.0	97.2
	10^-5	96.0	96.0
	10^-6	97.92	94.0
	10^-7	98.0	98.0
	10^-8	96.15	100.0
RTN	10^-4	98.0	98.0	98.0
	10^-5	98.0	98.0
	10^-6	97.96	96.0
	10^-7	100.0	98.0
	10^-8	96.15	100.0
DRTN	10^-4	100.0	100.0	99.2
	10^-5	98.04	100.0
	10^-6	98.0	98.0
	10^-7	100.0	100.0
	10^-8	100.0	98.0

Table 3. $X_{T}$ performance indicators of the experimental dataset in this network measurement
View table
View in Article
Table 3. $X_{T}$ performance indicators of the experimental dataset in this network measurement
Model Temperature/℃ Precision/% Recall/% Accuracy/%
DRTN 10^-4 100.0 100.0 98.8
10^-5 98.0 98.0
10^-6 98.0 98.0
10^-7 98.04 100.0
10^-8 100.0 98.0

Tools

Get Citation

Copy Citation Text

Fengtao HE, Qianqian WU, Jianlei ZHANG, Yi YANG, Juan ZHANG, Xinyu YAO, Weilin ZHAO. Underwater Turbulence Detection Technology Based on Convolutional Neural Networks[J]. Acta Photonica Sinica, 2024, 53(1): 0111004

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category:

Received: Jun. 2, 2023

Accepted: Sep. 6, 2023

Published Online: Feb. 1, 2024

The Author Email: Fengtao HE (hefengtao@xupt.edu.cn)

DOI:10.3788/gzxb20245301.0111004

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology

Table 1. Simulation parameters

Table 1. Simulation parameters

Table 2. XT performance indicators of different classification networks in simulated datasets

Table 2. XT performance indicators of different classification networks in simulated datasets

Table 3. XT performance indicators of the experimental dataset in this network measurement

Table 3. XT performance indicators of the experimental dataset in this network measurement

Table 2. $X_{T}$ performance indicators of different classification networks in simulated datasets

Table 2. $X_{T}$ performance indicators of different classification networks in simulated datasets

Table 3. $X_{T}$ performance indicators of the experimental dataset in this network measurement

Table 3. $X_{T}$ performance indicators of the experimental dataset in this network measurement