[1]赵志辉,赵瑞珍,岑翼刚,等.基于稀疏表示与线性回归的图像快速超分辨率重建[J].智能系统学报,2017,12(1):8-14.[doi:10.11992/tis.201603039]
　ZHAO Zhihui,ZHAO Ruizhen,CEN Yigang,et al.Rapid super-resolution image reconstruction based on sparse representation and linear regression[J].CAAI Transactions on Intelligent Systems,2017,12(1):8-14.[doi:10.11992/tis.201603039]
点击复制

基于稀疏表示与线性回归的图像快速超分辨率重建

《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷: 12 期数: 2017年第1期 页码: 8-14 栏目: 学术论文—机器感知与模式识别 出版日期: 2017-02-25
Title:
Rapid super-resolution image reconstruction based on sparse representation and linear regression
作者:
赵志辉1,2, 赵瑞珍1,2, 岑翼刚1,2, 张凤珍1,2
1. 北京交通大学 信息科学研究所, 北京 100044;
2. 北京交通大学 现代信息科学与网络技术北京市重点实验室, 北京 100044
Author(s):
ZHAO Zhihui1,2, ZHAO Ruizhen1,2, CEN Yigang1,2, ZHANG Fengzhen1,2
1. Institute of Information Science, Beijing Jiaotong University, Beijing 100044, China;
2. Beijing Key Laboratory of Advanced Information Science and Network Technology, Beijing Jiaotong University, Beijing 100044, China
关键词:
线性回归; 超分辨率; 字典训练; 稀疏表示; 图像重建; 特征训练; 子空间; 邻域嵌入
Keywords:
linear regression; super-resolution; dictionary learning; sparse representation; image reconstruction; feature learning; subspace; neighborhood embedding
分类号:
TP391.41
DOI:
10.11992/tis.201603039
摘要:
单幅图像超分辨率的目的是从一幅低分辨率的图像来重构出高分辨率的图像。基于稀疏表示和邻域嵌入的超分辨率图像重建方法使得重建图像质量有了极大的改善。但这些方法还很难应用到实际中，因为其重建图像的速度太慢或者需要调节复杂的参数。目前大多数的方法在图像重建的速度和质量两个方面很难有一个好的权衡。鉴于以上问题提出了一种基于线性回归的快速图像超分辨率重建算法，将稀疏表示和回归的方法有效地结合在一起。通过稀疏表示训练的字典，用一种新的方式将整个数据集划分为多个子空间，然后在每一类子空间中独立地学习高低分辨率图像之间的映射关系，最后通过选择相应的投影矩阵来重建出高分辨图像。实验结果表明，相比于其他方法，本文提出的算法无论在图像重建速度还是重建质量方面都取得了更好的超分辨率重建效果。
Abstract:
Single-image super-resolution aims at reconstructing a high-resolution image from a single low-resolution image. Recent methods relying on both neighborhood embedding and sparse coding have led to significant quality improvements. However, the application of these approaches is still practically difficult because they are either too slow or demand tedious parameter tweaks. In most of these methods, the speed and quality of image reconstruction are the two aspects that cannot be balanced easily. With regard to the abovementioned problems, this research proposed a rapid image super-resolution reconstruction algorithm based on linear regression, which effectively combined the sparse representation with the regression method. First, a dictionary was trained using the K-SVD algorithm based on training samples. Subsequently, the entire dataset was divided into a number of subspaces according to the atoms in the dictionary. Moreover, the mapping from low-to-high-resolution images can be independently obtained for each subspace. Finally, the high-resolution image was reconstructed by selecting the corresponding projection matrix. Experimental results demonstrate that both the image reconstruction quality and the speed of the proposed algorithm performed better than other widely used methods.
参考文献/References:
[1] BHAVSAR A V. Range image super-resolution via reconstruction of sparse range data[C]//Proceedings of the 2013 International Conference on Intelligent Systems and Signal Processing. Gujarat, India, 2013: 198-203.
[2] YANG Minchun, WANG Y C F. A self-learning approach to single image super-resolution[J]. IEEE transactions on multimedia, 2013, 15(3): 498-508.
[3] PARK S C, PARK M K, KANG M G. Super-resolution image reconstruction: a technical overview[J]. IEEE signal processing magazine, 2003, 20(3): 21-36.
[4] LI Xin, ORCHARD M T. New edge-directed interpolation[J]. IEEE transactions on image processing, 2001, 10(10): 1521-1527.
[5] FATTAL R. Image upsampling via imposed edge statistics[J]. ACM transactions on graphics (TOG), 2007, 26(3): 95.
[6] FREEMAN W T, JONES T R, PASZTOR E C. Example-based super-resolution[J]. IEEE computer graphics and applications, 2002, 22(2): 56-65.
[7] YANG C Y, YANG M H. Fast direct super-resolution by simple functions[C]//Proceedings of 2013 IEEE International Conference on Computer Vision. Sydney, Australia, 2013: 561-568.
[8] SUN Jian, XU Zongben, SHUM H Y, et al. Image super-resolution using gradient profile prior[C]//Proceedings of 2008 IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, USA, 2008: 1-8.
[9] DONG Chao, LOY C C, HE Kaiming, et al. Learning a deep convolutional network for image super-resolution[M]//FLEET D, PAJDLA T, SCHIELE B, et al. Computer Vision-ECCV 2014. Switzerland: Springer, 2014: 184-199.
[10] OLSHAUSEN B A, FIELD D J. Sparse coding with an overcomplete basis set: a strategy employed by V1[J]. Vision research, 1997, 37(23): 3311-3325.
[11] TIMOFTE R, DE V, VAN GOOL L. Anchored neighborhood regression for fast example-based super-resolution[C]//Proceedings of 2013 IEEE International Conference on Computer Vision. Sydney, Australia, 2013: 1920-1927.
[12] CHANG Hong, YEUNG D Y, XIONG Yimin. Super-resolution through neighbor embedding[C]//Proceedings of the 2004 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC, USA, 2004.
[13] YANG Jianchao, WRIGHT J, HUANG T S, et al. Image super-resolution via sparse representation[J]. IEEE transactions on image processing, 2010, 19(11): 2861-2873.
[14] ZEYDE R, ELAD M, PROTTER M. On single image scale-up using sparse-representations[M]//BOISSONNAT J D, CHENIN P, COHEN A, et al. Curves and Surfaces. Berlin Heidelberg: Springer, 2012: 711-730.
[15] AHARON M, ELAD M, BRUCKSTEIN A. rmK—SVD: an algorithm for designing overcomplete dictionaries for sparse representation[J]. IEEE transactions on signal processing, 2006, 54(11): 4311-4322.
[16] BEVILACQUA M, ROUMY A, GUILLEMOT C, et al. Low-complexity single-image super-resolution based on nonnegative neighbor embedding[C]//Proceedings of British Machine Vision Conference 2012. Guildford, Surrey, UK, 2012: 1-10.
[17] YANG C Y, MA C, YANG M H. Single-image super-resolution: a benchmark[M]//FLEET D, PAJDLA T, SCHIELE B, et al. Computer Vision-ECCV 2014. Switzerland: Springer, 2014: 372-386.
[18] TIMOFTE R, DE SMET V, VAN GOOL L. A+: adjusted anchored neighborhood regression for fast super-resolution[M]//CREMERS D, REID I, SAITO H, et al. Computer Vision—ACCV 2014. Switzerland: Springer, 2014: 111-126.
[19] SCHULTER S, LEISTNER C, BISCHOF H. Fast and accurate image upscaling with super-resolution forests[C]//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, MA, USA, 2015: 3791-3799.
相似文献/References:
[1]严菲,王晓栋.鲁棒的半监督多标签特征选择方法[J].智能系统学报,2019,14(4):812.[doi:10.11992/tis.201809017]
　YAN Fei,WANG Xiaodong.A robust, semi-supervised, and multi-label feature selection method[J].CAAI Transactions on Intelligent Systems,2019,14():812.[doi:10.11992/tis.201809017]
[2]王凡超,丁世飞.基于广泛激活深度残差网络的图像超分辨率重建[J].智能系统学报,2022,17(2):440.[doi:10.11992/tis.202106023]
　WANG Fanchao,DING Shifei.Image super-resolution reconstruction based on widely activated deep residual networks[J].CAAI Transactions on Intelligent Systems,2022,17():440.[doi:10.11992/tis.202106023]
[3]李海丰,李纪霖,王怀超,等.复杂机场道面外来异物高精度实时检测算法[J].智能系统学报,2023,18(3):525.[doi:10.11992/tis.202110014]
　LI Haifeng,LI Jilin,WANG Huaichao,et al.High-precision real-time detection algorithm for foreign object debris on complex airport pavements[J].CAAI Transactions on Intelligent Systems,2023,18():525.[doi:10.11992/tis.202110014]

备注/Memo

收稿日期:2016-3-19;改回日期:。
基金项目:国家自然科学基金项目（61272028，61572067）；国家“863”计划项目（2014AA015202）；广东省自然科学基金项目（2016A030313708）；北京市自然科学基金项目（4162050）.
作者简介:赵志辉,男,1990年生,硕士研究生,主要研究方向为稀疏表示与图像超分辨率;赵瑞珍,男,1975年生,教授,博士生导师,博士,主要研究方向为图像与信号处理算法、压缩感知与稀疏表示、信息感知域智能信息处理。主持参与国家自然科学基金、“863”计划等多项项目;岑翼刚,男,1978年生,教授,博士生导师,博士,主要研究方向为小波分析、压缩感知、图像处理。发表学术论文40余篇。
通讯作者:岑翼刚.E-mail:ygcen@bjtu.edu.cn.

更新日期/Last Update: 1900-01-01
Copyright © 《 智能系统学报》 编辑部
地址：（150001）黑龙江省哈尔滨市南岗区南通大街145-1号楼 电话：0451- 82534001、82518134　邮箱：tis@vip.sina.com