Uneven wrapped phase pattern denoising using a deep neural network

Jianming Li; Chen Tang; Min Xu; Zhenkun Lei

doi:10.1364/AO.461967

Applied Optics
Vol. 61,
Issue 24,
pp. 7150-7157
(2022)
•https://doi.org/10.1364/AO.461967

Uneven wrapped phase pattern denoising using a deep neural network

Jianming Li, Chen Tang, Min Xu, and Zhenkun Lei

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
Jianming Li, Chen Tang, Min Xu, and Zhenkun Lei, "Uneven wrapped phase pattern denoising using a deep neural network," Appl. Opt. 61, 7150-7157 (2022)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Check for updates

Abstract

The wrapped phase patterns obtained from an object composed of different materials have uneven gray values. In this paper, we improve the dilated-blocks-based deep convolution neural network (DBDNet) and build a new dataset for restoring the uneven gray values of uneven wrapped phase patterns as well as eliminating the speckle noise. In our method, we improve the structure of dilated blocks in DBDNet to enhance the ability of obtaining full scales of gray values and speckle noise information in the uneven phase patterns. We use the combined ${{\rm MS}\_{{\rm SSIM}}} + {\rm L1}$ loss function to improve the denoising and restoration performance of our method. We compare three representative networks ResNet-based, ADNet, and BRDNet in denoising with our proposed method. We test the three compared methods and our method on one group of computer-simulated and one group of experimentally obtained uneven noisy wrapped phase patterns from a dynamic measurement. We also conduct the ablation experiments on the improved model structure and the combined loss function used in our method. The denoising performance has been evaluated quantitatively and qualitatively. The denoising results demonstrate that our proposed method can reduce high speckle noise, restore the uneven gray values of wrapped phase patterns, and get better results than the compared methods.

Full Article | PDF Article

More Like This

DBDNet for denoising in ESPI wrapped phase patterns with high density and high speckle noise

Jianming Li, Chen Tang, Min Xu, Zirui Fan, and Zhenkun Lei
Appl. Opt. 60(32) 10070-10079 (2021)

UN-PUNet for phase unwrapping from a single uneven and noisy ESPI phase pattern

Hongxuan He, Chen Tang, Liao Zhang, Min Xu, and Zhenkun Lei
J. Opt. Soc. Am. A 40(10) 1969-1978 (2023)

Speckle suppression in holographic phase fringe patterns with different level noises based on FFDNet

Yun Liu, Xiaoqiang Wu, Qi Kang, Jiahao Gao, Mingxing Jiao, Junhong Xing, Xian Wang, and Hequn Li
Appl. Opt. 63(1) 77-84 (2024)

Previous Article Next Article

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (7)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (4)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (3)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Methods	Network	Input	Loss Function	Dilated Convolution	Skip Connection Number	Training Dataset
Yan et al. [22]	ResNet-based	Noisy sine/cosine fringe patterns	MSE	No	28	Speckle noise: $44, 000 - 40 \times 40$
Tian et al. [24]	ADNet	Gaussian/real noisy images	MSE	Yes	3	Gaussian noise: $1, 348, 480 - 50 \times 50$
		Gaussian/real noisy images				Real noise: $211, 600 - 50 \times 50$
Tian et al. [25]	BRDNet	Gaussian/real noisy images	MSE	Yes	3	Gaussian noise: $1, 231, 021 - 50 \times 50$
		Gaussian/real noisy images				Real noise: $423, 200 - 50 \times 50$
Javier et al. [26]	LRDUNet	Noisy fringe patterns	$L 1 + L 2$	No	4	Speckle noise: $192, 000 - 64 \times 64$
Li et al. [27]	DBDNet	Noisy phase patterns	MSE	Yes	18	Speckle noise: $15, 7000 - 50 \times 50$
Proposed method	DBDNet2	Noisy and uneven phase patterns	$M S_S S I M + L 1$	Yes	14	Speckle noise: $240, 260 (410 \times 586) - 50 \times 50$

Number of Interpolation Points	$n_{0}$	Number of Phase Images Generated	Total Number of Dataset 1
4	30	50	350
9	30	50
25	30	50
49	25	100
81	20	50
225	20	50

Network	Loss Function	PSNR	SSIM	MACs(G)
ResNet-based	MSE	16.8440	0.7712	482.953
ADNet	MSE	17.7993	0.8393	26.146
BRDNet	MSE	18.3621	0.8554	16.885
LRDUNet	$L 1 + L 2$	19.3058	0.9032	31.878
DBDNet2	$M S_S S I M + L 1$	20.8128	0.9239	21.901
Ablation Experiment
DBDNet	$M S_S S I M + L 1$	18.8819	0.8984	20.025
DBDNet2	MSE	19.1034	0.9078	21.901

Methods	Network	Input	Loss Function	Dilated Convolution	Skip Connection Number	Training Dataset
Yan et al. [22]	ResNet-based	Noisy sine/cosine fringe patterns	MSE	No	28	Speckle noise: $44, 000 - 40 \times 40$
Tian et al. [24]	ADNet	Gaussian/real noisy images	MSE	Yes	3	Gaussian noise: $1, 348, 480 - 50 \times 50$
		Gaussian/real noisy images				Real noise: $211, 600 - 50 \times 50$
Tian et al. [25]	BRDNet	Gaussian/real noisy images	MSE	Yes	3	Gaussian noise: $1, 231, 021 - 50 \times 50$
		Gaussian/real noisy images				Real noise: $423, 200 - 50 \times 50$
Javier et al. [26]	LRDUNet	Noisy fringe patterns	$L 1 + L 2$	No	4	Speckle noise: $192, 000 - 64 \times 64$
Li et al. [27]	DBDNet	Noisy phase patterns	MSE	Yes	18	Speckle noise: $15, 7000 - 50 \times 50$
Proposed method	DBDNet2	Noisy and uneven phase patterns	$M S_S S I M + L 1$	Yes	14	Speckle noise: $240, 260 (410 \times 586) - 50 \times 50$

Number of Interpolation Points	$n_{0}$	Number of Phase Images Generated	Total Number of Dataset 1
4	30	50	350
9	30	50
25	30	50
49	25	100
81	20	50
225	20	50

Abstract

Data availability

Cited By

Figures (7)

Tables (4)

Equations (3)

Applied Optics