基于卷积神经网络的超分辨率重建

被引:3
作者
张顺岚 [1 ]
曾儿孟 [2 ]
高宇 [2 ]
莫建文 [2 ]
机构
[1] 桂林电子科技大学机电工程学院
[2] 桂林电子科技大学信息与通信学院
关键词
深度学习; 图像超分辨率; 卷积神经网络; 稀疏表示; 网络参数;
D O I
10.16208/j.issn1000-7024.2017.11.034
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
分析超分辨率重建中基于卷积神经网络与基于稀疏表示方法的联系,讨论网络卷积核的作用,以及不同网络参数对重建效果的影响,设计一个权衡重建质量和结构复杂度的超分辨率重建卷积神经网络模型。实现低分辨率图像到高分辨率图像的映射,整个网络训练过程是端到端学习,具有全局优化的特点,避免基于稀疏表示方法的复杂特征提取和数据重组的操作,实验结果表明,该方法重建的图像质量在视觉效果和参数评价指标上有较大提高。
引用
收藏
页码:3080 / 3086
页数:7
相关论文
共 15 条
[1]  
A statistical prediction model based on sparse representations for single image super-resolution. Peleg T,Elad M. IEEE transactions on image processing:a publication of the IEEE Signal Processing Society . 2014
[2]  
Image Super-Resolution Via Sparse Representation. Jianchao Yang,Wright, J.,Huang, T.S.,Yi Ma. Image Processing, IEEE Transactions on . 2010
[3]   基于自适应权值滤波的深度图像超分辨率重建 [J].
杨宇翔 ;
曾毓 ;
何志伟 ;
高明煜 .
中国图象图形学报 , 2014, (08) :1210-1218
[4]  
Super-resolution through neighbor embedding. Chang H,Yeung D Y,Xiong Y. 2004 Computer Vision and Pattern Recognition . 2004
[5]  
Improving neural networks by preventing co-adaptation of feature detectors. Hinton G E,Srivastave N,Krizhevsky A,et al. Computer Science . 2012
[6]  
Convolutional sparse coding for image super-resolution. Gu S,Zuo W,Xie Q,et al. Proceedings of the IEEE International Conference on Computer Vision . 2015
[7]  
Learning a deep convolutional network for image super-resolution. DONG C,LOY C C,HE K,et al. Computer Vision-ECCV 2014 . 2014
[8]  
Efficient sparse coding algorithms. Lee H,Battle A,Raina R,et al. Neural Information Processing Systems . 2006
[9]  
Joint learning of multiple regressors for single image super-resolution. ZHANG K,WANG B,ZUO W,et al. IEEE Signal Processing Letters . 2016
[10]  
Anchored Neighborhood Regression for Fast Example-Based Super Resolution. Timofte R,De Smet V,Van Gool L. International Conference on Computer Vision(ICCV) . 2013