基于多尺度结构自相似性的单幅图像超分辨率算法

潘宗序; 禹晶; 胡少兴; 孙卫东

doi:10.3724/SP.J.1004.2014.00594

基于多尺度结构自相似性的单幅图像超分辨率算法

doi: 10.3724/SP.J.1004.2014.00594

1.
清华大学电子工程系北京 100084;
2.
北京航空航天大学机械工程与自动化学院北京 100083

基金项目:

国家自然科学基金（61171117），国家科技支撑计划项目（2012BAH31 B01），北京市教育委员会科技计划重点项目（KZ201310028035）资助

详细信息

作者简介:
胡少兴北京航空航天大学机械工程与自动化学院副教授.主要研究方向为三维激光扫描技术，图像处理，计算机视觉与模式识别.E-mail：husx@buaa.edu.cn

计量
- 文章访问数: 2847
- HTML全文浏览量: 238
- PDF下载量: 1516
- 被引次数: 0
出版历程
- 收稿日期: 2012-10-18
- 修回日期: 2013-02-18
- 刊出日期: 2014-04-20

Single Image Super Resolution Based on Multi-scale Structural Self-similarity

1.
Department of Electronic Engineering, Tsinghua University, Beijing 100084;
2.
School of Mechanical Engineering and Automation, Beijing University of Aeronautics and Astronautics, Beijing 100083

Funds:

Supported by National Natural Science Foundation of China (61171117), National Science and Technology Pillar Program of China (2012BAH31B01), and Key Project of the Science and Technology Development Program of Beijing Education Committee of China (KZ201310028035)

摘要

摘要: 多尺度结构自相似性是指同一幅图像中存在相同尺度或不同尺度的相似结构，这种多尺度图像结构自相似性广泛存在于遥感图像中.本文提出了一种基于多尺度结构自相似性的单幅图像超分辨率（Super resolution，SR）算法，该算法结合了压缩感知框架与图像结构自相似性，利用非局部方法和基于图像金字塔的K-SVD字典学习方法，将蕴含在相同尺度和不同尺度相似图像块中的附加信息在压缩感知的框架下加入到重构图像中.本文算法的优势在于，它仅借助于单幅低分辨率图像自身所蕴含的信息，实现了空间分辨率的提升.实验表明，与CSSS算法和ASDSAR算法相比，本文算法更有效地提升了遥感图像的空间分辨率.
- 超分辨率 /
- 结构自相似性 /
- 多尺度 /
- 压缩感知 /
- 非局部方法
Abstract: Multi-scale structural self-similarity refers to those similar structures either within the same scale or across different scales coming from the same image, which widely occur in remote sensing images. In this paper, we propose a single image super resolution (SR) method based on multi-scale structural self-similarity, which combines compressive sensing framework and structural self-similarity. In our method, the nonlocal and the pyramid-based K-SVD methods are used to add the extra information hidden in multi-scale structural self-similarity into the reconstructed image in the compressive sensing framework. The advantage of our method is that it only uses a single low-resolution image to promote spatial resolution by fully exploiting the extra information hidden in the image itself. Experimental results demonstrate that our method can improve spatial resolution more effectively compared with the CSSS and the ASDSAR methods.
- Super resolution (SR) /
- structural self-similarity /
- multi-scale /
- compressive sensing /
- nonlocal

HTML全文

参考文献(16)

[1]	Aly H A, Dubois E. Image up-sampling using total-variation regularization with a new observation model. IEEE Transactions on Image Processing, 2005, 14(10): 1647-1659
[2]	Babacan S D, Molina R, Katsaggelos A K. Total variation super resolution using a variational approach. In: Proceedings of the 15th IEEE International Conference on Image Processing. San Diego, USA: IEEE, 2008. 641-644
[3]	Sen P, Darabi S. Compressive image super-resolution. In: Proceedings of the 43rd Asilomar Conference on Signals, Systems and Computers. Pacific Grove, USA: IEEE, 2009. 1235-1242
[4]	Yang J C, Wright J, Huang T S, Ma Y. Image super-resolution via sparse representation. IEEE Transactions on Image Processing, 2010, 19(11): 2861-2873
[5]	Yang S Y, Sun F H, Wang M, Liu Z Z, Jiao L C. Novel super resolution restoration of remote sensing images based on compressive sensing and example patches-aided dictionary learning. In: Proceedings of the 2011 International Workshop on Multi-Platform/Multi-Sensor Remote Sensing and Mapping. Xiamen, China: IEEE, 2011. 1-6
[6]	Protter M, Elad M, Takeda H, Milanfar P. Generalizing the nonlocal-means to super-resolution reconstruction. IEEE Transactions on Image Processing, 2009, 18(1): 36-51
[7]	Mairal J, Bach F, Ponce J, Sapiro G, Zisserman A. Non-local sparse models for image restoration. In: Proceedings of the 12th International Conference on Computer Vision. Kyoto, Japan: IEEE, 2009. 2272-2279
[8]	Suetake N, Sakano M, Uchino E. Image super-resolution based on local self-similarity. Optical Review, 2008, 15(1): 26-30
[9]	Glasner D, Bagon S, Irani M. Super-resolution from a single image. In: Proceedings of the 12th International Conference on Computer Vision. Kyoto, Japan: IEEE, 2009. 349-356
[10]	Dong W S, Zhang L, Shi G M, Wu X L. Image deblurring and super-resolution by adaptive sparse domain selection and adaptive regularization. IEEE Transactions on Image Processing, 2011, 20(7): 1838-1857
[11]	Pan Zong-Xu, Huang Hui-Juan, Yu Jing, Hu Shao-Xing, Zhang Ai-Wu, Ma Hong-Bing, Sun Wei-Dong. Super-resolution method based on CS and structural self-similarity for remote sensing images. Signal Processing, 2012, 28(6): 859-872 (潘宗序, 黄慧娟, 禹晶, 胡少兴, 张爱武, 马洪兵, 孙卫东. 基于压缩感知与结构自相似性的遥感图像超分辨率方法. 信号处理, 2012, 28(6): 859-872)
[12]	Daubechies I, Defrise M, De Mol C. An iterative thresholding algorithm for linear inverse problems with a sparsity constraint. Communications on Pure and Applied Mathematics, 2004, 57(11): 1413-1457
[13]	Aharon M, Elad M, Bruckstein A. K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation. IEEE Transactions on Signal Processing, 2006, 54(11): 4311-4322
[14]	Elad M, Aharon M. Image denoising via sparse and redundant representations over learned dictionaries. IEEE Transactions on Image Processing, 2006, 15(12): 3736-3745
[15]	Moorthy A K, Bovik A C. A two-step framework for constructing blind image quality indices. IEEE Signal Processing Letters, 2010, 17(5): 513-516
[16]	Marziliano P, Dufaux F, Winkler S, Ebrahimi T. A no-reference perceptual blur metric. In: Proceedings of the 2002 International Conference on Image Processing. Rochester, USA: IEEE, 2002. III-57-III-60