Optics and Precision Engineering, Volume. 31, Issue 18, 2713(2023)
Uniform defocus blind deblurring based on deeper feature-based wiener deconvolution
Figures & Tables(11)
Fig. 1. Overview architecture of uniform defocusing deblurring network
Fig. 3. Architecture of deeper feature-based wiener deconvolution module
Fig. 9. Visual results of conductive particle detection after blurring the true blurred image
Table 1. Comparison of objective performance indicators of different algorithms
View tableView in ArticleTable 1. Comparison of objective performance indicators of different algorithms
Method DIV2K GOPRO MSE/10-3 PSNR/dB SSIM Time MSE/10-3 PSNR/dB SSIM Time SelfBlur［ 20 ］5.741 22.41 0.643 7 1.5 h 4.634 23.34 0.680 9 1.2 h DeepDeblur［ 15 ］1.439 28.42 0.798 4 0.04 s 0.753 31.23 0.895 5 0.03 s NAFNet［ 17 ］1.291 28.89 0.889 7 0.32 s 0.752 31.24 0.939 4 0.10 s SRN［ 16 ］1.183 29.27 0.839 7 0.05 s 0.625 32.04 0.913 1 0.05 s FWD［ 23 ］1.119 29.51 0.923 6 1.13 s 0.457 33.34 0.957 3 0.78 s UDBD（ours） 0.766 31.16 0.940 6 0.35 s 0.242 36.16 0.976 4 0.23 s
Table 2. Results of ablation experiment
View tableView in ArticleTable 2. Results of ablation experiment
DFWD FRM GOPRO PSNR/dB SSIM √ × 30.33 0.892 6 × √ 34.89 0.969 4 √ √ 36.16 0.976 4
Tools
Get Citation
Copy Citation Text
Chengxi WANG, Chen LUO, Jianghao ZHOU, Lang ZOU, Lei JIA. Uniform defocus blind deblurring based on deeper feature-based wiener deconvolution[J]. Optics and Precision Engineering, 2023, 31(18): 2713
Paper Information
Category: Information Sciences
Received: Nov. 21, 2022
Accepted: --
Published Online: Oct. 12, 2023
The Author Email: LUO Chen (chenluo@seu.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20