Image deblurring based on multiple local residual connection attention network

Table 1. Parameters configuration of residual module
View table
View in Article
Table 1. Parameters configuration of residual module
层卷积核通道数步长扩张因子
Conv1 3×3 64 1 1
Conv2 3×3 64 1 2
Conv3 3×3 64 1 3

Table 2. Parameters configuration of pairwise connection module
View table
View in Article
Table 2. Parameters configuration of pairwise connection module
层卷积核通道数步长扩张因子
Conv1_1 3x3 64 1 1
Conv1_2 3x3 64 1 2
Conv2_1 3x3 64 1 1
Conv2_2 3x3 64 1 3
Conv3_1 1x1 64 1 1

Table 3. Comparison of evaluation index results of different methods
View table
View in Article
Table 3. Comparison of evaluation index results of different methods
方法 Sun Nah Tao Zhang Xu 本文
PSRN/dB 24.64 29.08 30.26 31.20 31.23 31.83
SSIM 0.8429 0.9135 0.9342 0.9454 0.9455 0.9275
Size/MB 54.1 303.6 33.6 86.8 26.34 19.5
Runtime/s 12000 4300 1600 424 563 396

Table 4. Comparison of experimental results of different configuration algorithms on data set GoPro
View table
View in Article
Table 4. Comparison of experimental results of different configuration algorithms on data set GoPro
不同配置 PAM- RM- MLRC- RB PCM- 本文
PSNR/dB 28.86 29.38 29.56 29.73 29.76 31.83
SSIM 0.8932 0.9106 0.9131 0.9156 0.9147 0.9275

Tools

Get Citation

Copy Citation Text

Qingjiang CHEN, Qiaoying WANG. Image deblurring based on multiple local residual connection attention network[J]. Journal of Applied Optics, 2023, 44(2): 337

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites