Researching | Lens-less imaging via score-based generative model

Optics and Precision Engineering, Volume. 30, Issue 18, 2280(2022)

Lens-less imaging via score-based generative model

Chunhua WU... Hong PENG, Qiegen LIU, Wenbo WAN* and Yuhao WANG |Show fewer author(s)

Author Affiliations

School of Information Engineering， Nanchang University， Nanchang330031， China

show less

Abstract Get PDF(in Chinese)

Figures&Tables (13)

References (32)

Paper Information

Figures & Tables(13)

Fig. 1. Flow chart of lens-less imaging

Download full size

View in Article

Fig. 2. Generative model

Download full size

View in Article

Fig. 3. Forward and reverse time process of SDE

Download full size

View in Article

Fig. 4. Training and reconstruction flow chart of LSGM algorithm

Download full size

View in Article

Fig. 5. Visual comparison of reconstruction images on the LSUN-bedroom dataset

Download full size

View in Article

Fig. 6. Visual comparison of reconstruction images on the LSUN-church dataset

Download full size

View in Article

Fig. 7. Improvement of visual effects by LSGM algorithm with TV method

Download full size

View in Article

Fig. 8. Ablation experiments on the relationship between image quality and discretization steps under different σmax

Download full size

View in Article

Fig. 9. Experimental test of field of view

Download full size

View in Article

Fig. 10. Experimental test of the resolution

Download full size

View in Article

Table 1. Test results on LSUN-bedroom dataset
View table
View in Article
Table 1. Test results on LSUN-bedroom dataset
Method PSNR/dB SSIM
BP 7.78 0.23
CS 18.47 0.31
LSGM 26.10 0.64

Table 1. [in Chinese]

View in Article

Table 1. [in Chinese]

Algorithm 1 LSGM
Training
Dataset： Object image dataset Output： $S_{θ} (O, t) ≃ \nabla_{O} l o g p_{t} (O)$
Reconstruction
Setting： Score network $S_{θ}$ ， Noise iteration step $\{ε_{i}\}$ ， Noise scale $\{σ_{i}\}$ ，Number of discretization steps for the reverse-time SDE $N$ ， Number of corrector steps $M$ 1： $O_{N} \sim N (0, σ_{m a x}^{2} I)$ 2： For $i = N - 1$ to 0 do （Outer loop）
3： Update $O_{i}$ via Eq. （12） and （13）【Predictor】
4： For $j = 1$ to $M$ do （Inner loop）
5： Update $O_{i}$ via Eq. （14） and （15）【Corrector】
6： End for 7： End for 8： Return $O_{0}$

Table 2. Test results on LSUN-church dataset
View table
View in Article
Table 2. Test results on LSUN-church dataset
Method PSNR/dB SSIM
BP 7.81 0.22
CS 17.79 0.29
LSGM 24.47 0.65

Tools

Get Citation

Copy Citation Text

Chunhua WU, Hong PENG, Qiegen LIU, Wenbo WAN, Yuhao WANG. Lens-less imaging via score-based generative model[J]. Optics and Precision Engineering, 2022, 30(18): 2280

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Information Sciences

Received: Mar. 25, 2022

Accepted: --

Published Online: Oct. 27, 2022

The Author Email: WAN Wenbo (wanwenbo@ncu.edu.cn)

DOI:10.37188/OPE.20223018.2280

Topics

laser devices and laser physics

Lasers and Laser Optics

laser manufacturing

Instrumentation, Measurement and Metrology