结合大气散射模型的生成对抗网络去雾算法
|
|
屠杭垚,王万良,陈嘉诚,李国庆,吴菲
|
Dehazing algorithm combined with atmospheric scattering model based on generative adversarial network
|
|
Hang-yao TU,Wan-liang WANG,Jia-chen CHEN,Guo-qing LI,Fei WU
|
|
| 表 4 不同损失函数下图像指标的对比 |
| Tab.4 Comparison of image indicator in different loss functions |
|
| 图像1 | PSNR | SSIM | | 图像2 | PSNR | SSIM | | 图像3 | PSNR | SSIM | | CGAN | 15.010 2 | 0.522 85 | | CGAN | 17.133 0 | 0.555 39 | | CGAN | 16.321 9 | 0.585 44 | | CGAN+L1 | 24.739 5 | 0.749 23 | CGAN+L1 | 25.648 2 | 0.782 92 | CGAN+L1 | 24.770 9 | 0.748 43 | | CGAN+雾霾损失 | 29.502 0 | 0.805 62 | CGAN+雾霾损失 | 30.770 9 | 0.817 03 | CGAN+雾霾损失 | 28.502 0 | 0.805 62 | | CGAN+L1+雾霾损失 | 33.927 2 | 0.907 04 | CGAN+L1+雾霾损失 | 35. 689 8 | 0.902 60 | CGAN+L1+雾霾损失 | 32.180 9 | 0.894 32 |
|
|
|