基于生成对抗网络的梯度引导太阳斑点图像去模糊方法
Solar speckle image deblurring method with gradient guidance based on generative adversarial network
查看参考文献26篇
|
文摘
|
针对云南天文台拍摄的高度模糊的太阳斑点图像采用现有深度学习算法恢复难度大、高频信息难以重建等问题,提出了一种基于生成对抗网络(GAN)与梯度信息联合的去模糊方法来重建太阳斑点图,并很好地恢复出图像的高频信息。该方法由一个生成器与两个鉴别器构成:首先,生成器采用特征金字塔网络(FPN)框架来获取图像多尺度特征,再将这些特征分层次输入梯度分支以梯度图的形式捕获更小的局部特征;然后,联合梯度分支结果与FPN结果共同重建出具有高频信息的太阳斑点图像;其次,在常规对抗鉴别器的基础上,增加了一个鉴别器用于保证由梯度分支产生的梯度图更加真实;最后,引入一个包括像素内容损失、感知损失和对抗损失的联合训练损失来引导模型进行太阳斑点图像高分辨率重建。实验结果表明,进行图像预处理后的所提方法与现有的深度学习去模糊方法相比,高频信息恢复能力更强,峰值信噪比(PSNR)和结构相似性(SSIM)指标均有显著提高,分别达到27.801 0 dB与0.851 0,能够满足太阳观测图像高分辨率重建的需要。 |
|
其他语种文摘
|
With the existing deep learning algorithms,it is difficult to restore the highly blurred solar speckle images taken by Yunnan Observatories,and it is difficult to reconstruct the high-frequency information of images. In order to solve the problems,a deblurring method for restoring the solar speckle images and recovering the high-frequency information of images based on Generative Adversarial Network(GAN)and gradient information was proposed. The proposed method was consisted of one generator and two discriminators. Firstly,the image multi-scale features were obtained by the generator with the Feature Pyramid Network(FPN)framework,and these features were input into the gradient branch hierarchically to capture the smaller details in the form of gradient map,and the solar speckle image with high-frequency information was reconstructed by combining the gradient branch results and the FPN results. Then,based on the conventional adversarial discriminator,another discriminator was added to ensure the gradient map generated by the gradient branch more realistic. Finally,a joint training loss including pixel content loss,perceptual loss and adversarial loss was introduced to guide the model to perform high-resolution reconstruction of solar speckle images. Experimental results show that,compared with the existing deep learning deblurring method,the proposed method with image preprocessing has stronger ability to recover the high-frequency information,and significantly improves the Peak Signal-to-Noise Ratio(PSNR)and Structural SIMilarity (SSIM)indicators,reaching 27.801 0 dB and 0.851 0 respectively. The proposed method can meet the needs for highresolution reconstruction of solar observation images. |
|
来源
|
计算机应用
,2021,41(11):3345-3352 【扩展库】
|
|
DOI
|
10.11772/j.issn.1001-9081.2020121898
|
|
关键词
|
去模糊
;
生成对抗网络
;
梯度引导
;
局部细节
;
太阳斑点
|
|
地址
|
1.
云南大学信息学院, 昆明, 650500
2.
中国科学院云南天文台, 昆明, 650216
|
|
语种
|
中文 |
|
文献类型
|
研究性论文 |
|
ISSN
|
1001-9081 |
|
学科
|
自动化技术、计算机技术 |
|
基金
|
国家自然科学基金
;
云南省高校科技创新团队支持项目
|
|
文献收藏号
|
CSCD:7110554
|
参考文献 共
26
共2页
|
|
1.
张兰强.
太阳高分辨率力成像多层共轭自适应光学技术研究,2014:20-28
|
CSCD被引
1
次
|
|
|
|
|
2.
刘忠.
天文图象高分辨重建及空域性质研究,2003:25-39
|
CSCD被引
1
次
|
|
|
|
|
3.
邱耀辉. 天文图像空域重建新方法:迭代位移叠加法.
光学学报,2001,21(2):186-191
|
CSCD被引
10
次
|
|
|
|
|
4.
向永源.
太阳高分辨高速重建算法的研究,2016:19-22
|
CSCD被引
1
次
|
|
|
|
|
5.
Lin T Y. Feature pyramid networks for object detection.
Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition,2017:936-944
|
CSCD被引
95
次
|
|
|
|
|
6.
Ma C. Structure-preserving super resolution with gradient guidance.
Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition,2020:7766-7775
|
CSCD被引
3
次
|
|
|
|
|
7.
Johnson J. Perceptual losses for real-time style transfer and super-resolution.
Proceedings of the 2016 European Conference on Computer Vision,LNCS 9906,2016:1604-1613
|
CSCD被引
1
次
|
|
|
|
|
8.
Gu J J. Blind super-resolution with iterative kernel correction.
Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition,2019:1604-1613
|
CSCD被引
3
次
|
|
|
|
|
9.
Guo Y. Closed-loop matters:dual regression networks for single image super-resolution.
Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition,2020:5406-5415
|
CSCD被引
3
次
|
|
|
|
|
10.
Kaufman A. Deblurring using analysis synthesis networks pair.
Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition,2020:5810-5819
|
CSCD被引
1
次
|
|
|
|
|
11.
Yang F Z. Learning texture transformer network for image super-resolution.
Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition,2020:5790-5799
|
CSCD被引
2
次
|
|
|
|
|
12.
Goodfellow I J. Generative adversarial networks.
Proceedings of the 27th International Conference on Neural Information Processing Systems,2014:2672-2680
|
CSCD被引
362
次
|
|
|
|
|
13.
Radford A.
Unsupervised representation learning with deep convolutional generative adversarial networks,2016
|
CSCD被引
72
次
|
|
|
|
|
14.
Zhu J Y. Unpaired image-to-image translation using cycle-consistent adversarial networks.
Proceedings of the 2017 IEEE International Conference on Computer Vision,2017:2242-2251
|
CSCD被引
29
次
|
|
|
|
|
15.
Engin D. Cycle-dehaze:enhanced CycleGAN for single image dehazing.
Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops,2018:938-946
|
CSCD被引
2
次
|
|
|
|
|
16.
Kupyn O. DeblurGAN-v2: deblurring (orders-of-magnitude) faster and better.
Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision,2019:8877-8886
|
CSCD被引
5
次
|
|
|
|
|
17.
Jolicoeur-Martineau A.
The relativistic discriminator:a key element missing from standard GAN,2020
|
CSCD被引
1
次
|
|
|
|
|
18.
Gulrajani I. Improved training of Wasserstein GANs.
Proceedings of the 31st International Conference on Neural Information Processing Systems,2017:5769-5779
|
CSCD被引
71
次
|
|
|
|
|
19.
Ren Y H. Reconstruction of singleframe solar speckle image with cycle consistency loss and perceptual loss.
Proceedings of the 2019 6th International Conference on Information Science and Control Engineering,2019:439-443
|
CSCD被引
1
次
|
|
|
|
|
20.
Jia P. Solar image restoration with the CycleGAN based on multi-fractal properties of texture features.
The Astrophysical Journal Letters,2019,881(2):Article No. L30
|
CSCD被引
1
次
|
|
|
|
|
了解气象卫星更多信息，请访问