Infrared and Laser Engineering, Volume. 49, Issue 10, 20200267(2020)
Aberration correction for flow velocity measurements using deep convolutional neural networks
Figures & Tables(8)
Fig. 3. Measurement result of Zernike wavefront (First 10 order Zernike coefficients amplitude spectrum of fluctuating air-water interface induced wavefront aberration)
Fig. 5. Aberration correction of single frame PIV image. (a) Input wavefront aberration; (b) Original PIV image; (c) Distorted PIV image; (d) Corrected PIV image; (e) Restored PIV image by ALM
Fig. 6. PIV measurement results from (a) Ideal images; (b) Distorted images; (c) Corrected images
Table 1. [in Chinese]
View tableView in ArticleTable 1. [in Chinese]
Distorted PIV image Corrected PIV image Restored PIV image by ALM MSE 0.064 0.00062 0.056 PSNR 12.79 32.16 15.14 SSIM 0.37 0.73 0.39
Table 2. [in Chinese]
View tableView in ArticleTable 2. [in Chinese]
$\sigma $ $\sigma {\rm{/}}\bar v$ Reduction Ideal PIV 0.736 12.04 % — Distorted PIV 5.003 88.07 % 0 Corrected PIV 0.846 14.23 % 97 %
Tools
Get Citation
Copy Citation Text
Zeyu Gao, Xinyang Li, Hongwei Ye. Aberration correction for flow velocity measurements using deep convolutional neural networks[J]. Infrared and Laser Engineering, 2020, 49(10): 20200267
Paper Information
Category: Adaptive optics
Received: Jun. 18, 2020
Accepted: --
Published Online: Jul. 6, 2021
The Author Email:
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20