Page 319 - 《软件学报》2021年第10期

张学锋 等:基于双注意力残差循环单幅图像去雨集成网络 3291

的思想提出了门控网络,自主学习和分配各个阶段的权重,将每个阶段的输出结果与门控网络对应的权重相乘
相加.通过仿真和真实的雨数据实验,结果表明:本文的方法相对于文中其他图像去雨方法,在雨痕去除和纹理
细节保留与恢复上均取得了明显的提升.在未来的工作中,我们计划在非匹配真实的雨天情况下实现单幅图像
的去雨,因为仿真雨数据集不能很好地模拟实际复杂雨天的场景,希望通过不匹配的网络学习方式,达到真实情
景下的单幅图像去雨任务.

References:
[1] Li Y, Tan RT, Guo XJ, Lu J, Brown MS. Rain streak removal using layer priors. In: Proc. of the IEEE Conf. on Computer Vision
and Pattern Recognition. 2016. 27362744.
[2] Gu SH, Meng DY, Zuo WM, Zhang L. Joint convolutional analysis and synthesis sparse representation for single image layer
separation. In: Proc. of the IEEE Int’l Conf. on Computer Vision. 2017. 17081716.
[3] Kang LW, Lin CW, Fu YH. Automatic single-image-based rain streaks removal via image decomposition. IEEE Trans. on Image
Processing, 2012,21(4):17421755.
[4] Luo Y, Xu Y, Ji H. Removing rain from a single image via discriminative sparse coding. In: Proc. of the IEEE Int’l Conf. on
Computer Vision. 2015. 33973405.
[5] Wang YL, Liu SC, Chen C, Zeng B. A hierarchical approach for rain or snow removing in a single color image. IEEE Trans. on
Image Processing, 2017,26(8):39363950.
[6] Dong C, Loy CC, He KM, Tang XO. Image super-resolution using deep convolutional networks. IEEE Trans. on Pattern Analysis
and Machine Intelligence, 2015,38(2):295307.
[7] Zhang K, Zuo WM, Chen YJ, Meng DY, Zhang L. Beyond a gaussian denoiser: Residual learning of deep CNN for image
denoising. IEEE Trans. on Image Processing, 2017,26(7):31423155.
[8] Tai Y, Yang J, Liu XM. Image super-resolution via deep recursive residual network. In: Proc. of the IEEE Conf. on Computer
Vision and Pattern Recognition. 2017. 31473155.
[9] Fu XY, Huang JB, Ding XH, Liao YH, Paisley J. Clearing the skies: A deep network architecture for single-image rain removal.
IEEE Trans. on Image Processing, 2017,26(6):29442956.
[10] Zhang H, Patel VM. Density-aware single image deraining using a multi-stream dense network. In: Proc. of the IEEE Conf. on
Computer Vision. 2018. 695704.
[11] Fu XY, Huang JB, Zeng DL, Huang Y, Ding XH, Paisley J. Removing rain from single images via a deep detail network. In: Proc.
of the IEEE Conf. on Computer Vision and Pattern Recognition. 2017. 38553863.
[12] Yang WH, Tan RT, Feng JS, Liu JY, Guo ZM, Yan SC. Deep joint rain detection and removal from a single image. In: Proc. of the
IEEE Conf. on Computer Visionand Pattern Recognition. 2017. 13571366.
[13] Wang TY, Yang X, Xu K, Chen SZ, Zhang Q, Lau RW. Spatial attentive single-image deraining with a high quality real rain
dataset. In: Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition. 2019. 122701229.
[14] Yang YZ, Lu H. Single image deraining using a recurrent multi-scale aggregation and enhancement network. In: Proc. of the IEEE
Int’l Conf. on Multimedia and Expo. 2019. 13781383.
[15] Li X, Wu JL, Lin ZC, Liu H, Zha HB. Recurrent squeeze-and-excitation context aggregation net for single image deraining. In:
Proc. of the European Conf. on Computer Vision. 2018. 254269.
[16] Ren DW, Zuo WM, Hu QH, Zhu PF, Meng DY. Progressive image deraining networks: A better and simpler baseline. In: Proc. of
the IEEE Conf. on Computer Vision and Pattern Recognition. 2019. 39373946.
[17] Li JJ. Research on single image rain removal based on convolutional neural network [MS. Thesis]. Maanshan: Anhui University of
Technology, 2020 (in Chinese with English abstract).
[18] Wang ZY, Ji SW. Smoothed dilated convolutions for improved dense prediction. In: Proc. of the 24th ACMSIGKDD Int’l Conf. on
Knowledge Discovery & Data Mining. 2018. 24862495.
[19] Shi XJ, Chen ZR, Wang H, Yeung DY, Wong WK, Woo WC. Convolutional LSTM network: A machine learning approach for
precipitation nowcasting. In: Advances in Neural Information Processing Systems. 2015. 802810.