有去噪和尺度放大的单幅图像超分辨率重建

doi:10.7523/j.issn.2095-6134.2016.05.004

中国科学院大学学报 ›› 2016, Vol. 33 ›› Issue (5): 596-603.DOI: 10.7523/j.issn.2095-6134.2016.05.004

有去噪和尺度放大的单幅图像超分辨率重建

刘晓¹, 郭田德^1,2, 韩丛英^1,2, 李明强¹

1 中国科学院大学数学科学学院, 北京 100049;
2 中国科学院大数据挖掘与知识管理重点实验室, 北京 100190

收稿日期:2015-12-25 修回日期:2016-03-15 发布日期:2016-09-15
通讯作者: 韩丛英
基金资助:
Supported by the National Nature Science Foundation of China (11331012,11571014,71271204)

Super resolution reconstruction of single image with denoising and upscaling

LIU Xiao¹, GUO Tiande^1,2, HAN Congying^1,2, LI Mingqiang¹

1 School of Mathematical Sciences, University of Chinese Academy of Sciences, Beijing 100049, China;
2 Key Laboratory of Big Data Mining and Knowledge Management, Chinese Academy of Sciences, Beijing 100190, Chin

Received:2015-12-25 Revised:2016-03-15 Published:2016-09-15
Supported by:
Supported by the National Nature Science Foundation of China (11331012,11571014,71271204)

摘要/Abstract

摘要：

采用非局部均值，全变差正则化和稀疏编码，提出一种新的单幅图像超分辨率重建方法.首先，用非局部均值方法对低分辨率图像进行去噪处理，可保证几何结构的一致性.然后，将去噪后的低分辨率图像代入重建模型得到高分辨率图像的低频分量，而高分辨率图像的高频分量则通过学习的方法在数据库中查找得到.将得到的低频分量与高频分量相加得到初始的高分辨率图像的恢复结果.最后对初始的恢复图像进行一些全局优化即可得到最终恢复的高分辨率图像结果.实验结果证明该方法得到的高分辨率图像结果具有鲁棒性且更自然.

关键词: 超分辨率, 非局部均值, 全变差, 稀疏编码

Abstract:

In this study, a novel approach to single image resolution reconstruction is proposed based on nonlocal means, total variation-regularization, and sparse coding. Firstly, low-resolution (LR) image is denoised by nonlocal means which preserves geometric structure well. Then, high-resolution (HR) regularization-based component is obtained from up-scaling the LR image by using a reconstruction model. Image generation process is combined with total-variation regularization so that new image maintains part of the sharpness of the edges and some details. Meanwhile, the HR learning-based component is reconstructed by exploring the sparse coding which explores the co-occurrence relationship between LR training patches and their corresponding HR high-frequency patches. The regularization-based component and the learning-based component are combined to obtain an initial HR image. Finally, the global reconstruction constraint is applied to the initial image for making the final HR image natural. Experimental results show that our method is natural and robust.

Key words: super resolution, nonlocal means, total variation, sparse coding

中图分类号:

O221.7
O235

刘晓, 郭田德, 韩丛英, 李明强. 有去噪和尺度放大的单幅图像超分辨率重建[J]. 中国科学院大学学报, 2016, 33(5): 596-603.

LIU Xiao, GUO Tiande, HAN Congying, LI Mingqiang. Super resolution reconstruction of single image with denoising and upscaling[J]. , 2016, 33(5): 596-603.

参考文献

[1] Park S C, Park M K, Kang M G. Super-resolution image reconstruction:a technical overview[J]. Signal Processing Magazine, IEEE, 2003, 20(3):21-36.
[2] Li X, Orchard M T. New edge-directed interpolation[J]. Image Processing, IEEE Transactions on, 2001, 10(10):1521-1527.
[3] Takeda H, Farsiu S, Milanfar P. Kernel regression for image processing and reconstruction[J]. Image Processing, IEEE Transactions on, 2007, 16(2):349-366.
[4] Chang H, Yeung D Y, Xiong Y. Super-resolution through neighbor embedding[C]//Computer Vision and Pattern Recognition, 2004. Proceedings of the 2004 IEEE Computer Society Conference on. IEEE, 2004, 1:275-282.
[5] Farsiu S, Robinson M D, Elad M, et al. Fast and robust multiframe super resolution[J]. Image processing, IEEE Transactions on, 2004, 13(10):1327-1344.
[6] Buades A, Coll B, Morel J M. A non-local algorithm for image denoising[C]//Computer Vision and Pattern Recognition, 2005. IEEE Computer Society Conference on. IEEE, 2005, 2:60-65.
[7] Freeman W T, Jones T R, Pasztor E C. Example-based super-resolution[J]. Computer Graphics and Applications, IEEE, 2002, 22(2):56-65.
[8] Yang J, Wright J, Huang T, et al. Image super-resolution as sparse representation of raw image patches[C]//Computer Vision and Pattern Recognition, 2008. IEEE Conference on. IEEE, 2008:1-8.
[9] Glasner D, Bagon S, Irani M. Super-resolution from a single image[C]//Computer Vision, 2009 IEEE 12th International Conference on. IEEE, 2009:349-356.
[10] Li J, Gong W, Li W, et al. Single-image super-resolution reconstruction based on global non-zero gradient penalty and non-local laplacian sparse coding[J]. Digital Signal Processing, 2014, 26:101-112.
[11] Lin Z, Shum H Y. Fundamental limits of reconstruction-based superresolution algorithms under local translation[J]. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 2004, 26(1):83-97.
[12] Protter M, Elad M, Takeda H, et al. Generalizing the nonlocal-means to super-resolution reconstruction[J]. Image Processing, IEEE Transactions on, 2009, 18(1):36-51.
[13] Yu J, Gao X, Tao D, et al. A unified learning framework for single image super-resolution[J]. Neural Networks and Learning Systems, IEEE Transactions on, 2014, 25(4):780-792.
[14] Nguyen N, Milanfar P, Golub G. A computationally efficient superresolution image reconstruction algorithm[J]. Image Processing, IEEE Transactions on, 2001, 10(4):573-583.
[15] Rudin L I, Osher S, Fatemi E. Nonlinear total variation based noise removal algorithms[J]. Physica D:Nonlinear Phenomena, 1992, 60(1):259-268.
[16] Beck A, Teboulle M. Fast gradient-based algorithms for constrained total variation image denoising and deblurring problems[J]. Image Processing, IEEE Transactions on, 2009, 18(11):2419-2434.
[17] Freeman W T, Pasztor E C, Carmichael O T. Learning low-level vision[J]. International Journal of Computer Vision, 2000, 40(1):25-47.
[18] Sun J, Zheng N N, Tao H, et al. Image hallucination with primal sketch priors[C]//Computer Vision and Pattern Recognition, 2003. Proceedings. 2003 IEEE Computer Society Conference on. IEEE, 2003, 2:729-736.
[19] Martin D, Fowlkes C, Tal D, et al. A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics[C]//Computer Vision, 2001. Proceedings Eighth IEEE International Conference on. IEEE, 2001, 2:416-423.

有去噪和尺度放大的单幅图像超分辨率重建

Super resolution reconstruction of single image with denoising and upscaling

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

本文评价

访问统计

联系我们

[1]	常敬明, 金铭, 曾江源, 陈锟山. 随机介质场景下的时间反转成像效果分析[J]. 中国科学院大学学报, 2018, 35(1): 59-65.
[2]	姚振杰, 易卫东. 一种用于车牌识别的图像超分辨率算法[J]. 中国科学院大学学报, 2013, 30(1): 137-143.
[3]	袁建华; 殷学民; 邹谋炎. 一种有效的超分辨率重建计算方法[J]. 中国科学院大学学报, 2006, 23(4): 514-519.