基于耦合过完备字典的红外云图超分辨率方法

石大维; 符冉迪; 金炜

doi:doi:10.3969/j.issn.1003-501x.2014.04.011

光电工程, 2014, 41 (4): 69, 网络出版: 2014-04-09

基于耦合过完备字典的红外云图超分辨率方法

A Method of Infrared Nephogram Super-resolution Based on Coupled Over-completed Dictionary

论文大纲

石大维 ^*符冉迪金炜

作者单位

宁波大学信息科学与工程学院, 浙江宁波 315211

红外云图超分辨率耦合字典学习稀疏表示 infrared nephogram super-resolution coupled dictionary learning sparse representation

摘要

针对红外云图分辨率低的问题, 提出一种基于耦合过完备字典的超分辨率方法。在分析红外云图成像退化模型的基础上, 建立了采用稀疏表示理论的超分辨率重构框架, 首先随机抽取大量高、低分辨率云图的图像块, 组成训练样本, 经过字典学习获取针对高、低分辨率云图块的两个字典Dh 和Dl, 为保证对应的高、低分辨率云图块关于各自的字典具有相似的稀疏表示, 提出一种耦合字典学习算法, 该算法改变了字典对的更新策略, 通过在每一步迭代中交替优化Dh 和Dl, 得到耦合的过完备字典对; 最后对输入的低分辨率红外云图, 采用最优正交匹配追踪算法(Optimized Orthogonal Matching Pursuit Algorithm, OOMP), 得到满足重构约束的高分辨率云图。实验结果表明, 本文方法与其他方法相比, 红外云图重构质量有较为明显的改善, 而且比同类方法具有更高的计算效率。

Abstract

For the problem of low-resolution of infrared nephogram, a method of infrared nephogram super-resolution using coupled over-completed dictionary learning is presented. Based on the analysis of infrared nephogram degradation model, a super-resolution reconstruction framework is built with the sparse representation theory. First, consist a training sample by random sampling a large number of high-resolution and low-resolution sample nephogram patches. Then, construct a couple of dictionaries Dh and Dl by dictionary training, to ensure that the corresponding high-resolution and low-resolution nephogram patches have a similar sparse representation on their dictionaries. We propose a coupled dictionary training method to change the strategy of dictionary update, and the coupled dictionaries were obtained by alternatively optimizing the Dh and Dl in each step of iterative. Finally, for the inputting low-resolution infrared nephogram, Optimized Orthogonal Matching Pursuit (OOMP) algorithm is used to achieve the high-resolution infrared nephogram which satisfies the reconstruction constraint. Numerous experiments show that the proposed algorithm can get a higher quality reconstructed infrared nephogram. Moreover, the algorithm efficiency is more effective than the other dictionary-learning algorithms.

参考文献

[1] Georgiev C G, Kozinarova G. Usefulness of satellite water vapour imagery in forecasting strong convection: A flash-flood case study [J]. Atmospheric Research(S0169-8095), 2009: 93(1/3): 295-303.

[2] Freeman T, Jones T R, Pasztor E C. Example based superresolution [J]. IEEE Computer Graphics and Applications (S0272-1716), 2002, 22(2): 56-65.

[3] Chang H, Yeung D, Xiong Y M. Super-resolution through neighbor embedding [C]// Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Washington, DC, USA, Jun 27-Jul 2, 2004, 1: 275-282.

[4] Ni K, Kumar S, Vasconchlos N, et al. Single image superresolution based on support vector regression [C]// Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, Toulouse, France, May 14-19, 2006, 2: 601-604.

[5] Yang J C, Wright J, Huang T S, et al. Image Super-Resolution via Sparse Representation [J]. IEEE Transactions on Image Processing(S1057-7149), 2010, 19(11): 2861-2873.

[6] Engan K, Aase S O. Method of optimal directions for frame design [C]// Proceedings of IEEE International Conference on Acoustics, Speech and Signal Proceeding, Phoenix, A Z, 1999, 5: 2443-2446.

[7] Aharon M, Elad M, Bruckstein A. K-SVD: An algorithm for designing overcomplete dictionaries for sparse representation [J]. IEEE Transactions on Signal Process(S1053-587X), 2006, 54(11): 4311–4322.

[8] 路锦正, 张启衡, 徐智勇, 等. 超完备稀疏表示的图像超分辨率重构方法 [J]. 系统工程与电子技术, 2012, 34(2): 403–408.

LU Jinzheng, ZHANG Qiheng, XU Zhiyong, et al. Image super-resolution algorithm using over-complete sparse representation [J]. System Engineering and Electronics, 2012, 34(2): 403-408.

[9] Rebollo-Neira L , Lowe D. Optimized orthogonal matching pursuit approach [J]. IEEE Singal Processing Letters (S1070-9908), 2002, 9(4): 137-140.

[10] WANG Baobin, WANG Yuxia. Some properties relating to stochastic gradient descent methods [J]. Journal of Mathematic (S0255-7797), 2011, 31(6): 1041-1044.

[11] DONG Weisheng, ZHANG Lei, SHI Guangming, et al. Image deblurring and super-resolution by adaptive sparse domain selection and adaptive regularization [J]. IEEE Transactions on Image Processing(S1057-7149), 2011, 20(7): 1838-1857.

[12] Mairal J, Bach F, Ponce F, et al. Online learning for matrix factorization and sparse coding [J]. Journal of Machine Learning Research(S1532-4435), 2011, 11(1): 19-60.

石大维, 符冉迪, 金炜. 基于耦合过完备字典的红外云图超分辨率方法[J]. 光电工程, 2014, 41(4): 69. SHI Dawei, FU Randi, JIN Wei. A Method of Infrared Nephogram Super-resolution Based on Coupled Over-completed Dictionary[J]. Opto-Electronic Engineering, 2014, 41(4): 69.

基于耦合过完备字典的红外云图超分辨率方法

关于本站 Cookie 的使用提示

全站搜索

基于耦合过完备字典的红外云图超分辨率方法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索