超分辨率算法研究综述

被引:34
作者
浦剑 [1 ]
张军平 [1 ]
黄华 [2 ]
机构
[1] 复旦大学计算机科学技术学院
[2] 西安交通大学电信学院
关键词
图像处理; 超分辨率; 邻域嵌入; 图像重建;
D O I
暂无
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
图像超分辨率是指利用一幅或多幅低分辨率图像,运用相应的算法来获得一幅清晰的高分辨率图像.然而,传统的基于插值和重建的方法已很难获得进一步的突破.近年来出现的基于学习的方法为超分辨率的发展重新注入了活力.通过回顾插值、重建和学习这3个层面的超分辨率算法,分析了超分辨率技术的以往研究和最新进展,着重讨论了各算法在还原质量、通用能力等方面所存在的问题,并对未来超分辨率技术的发展作了一些展望.
引用
收藏
页码:27 / 32
页数:6
相关论文
共 51 条
[1]   Hallucinating multiple occluded face images of different resolutions [J].
Jia, Kui ;
Gong, Shaogang .
PATTERN RECOGNITION LETTERS, 2006, 27 (15) :1768-1775
[2]  
On ambiguities in super-resolutionmodeling. Zhaozhong Wang,Feihu Qi. IEEE Signal Processing Letters . 2004
[3]  
Multiframe Image Restoration and Registration. TSAI R Y,HUANG T S. Advances in Computers . 1984
[4]  
Generalized interpolation and its ap-plication in super-resolution imaging. Rajan D,Chaudhuri S. Image and Vision Computing . 2001
[5]  
Super-resolution Enhancement of Video. Christopher M. Proceedings of the9th International Conference on Artificial Intelligence and Statistics . 2003
[6]  
Improving resolution byimage registration. Irani M,Peleg S. CV-GIP:Graphical Models andImage Processing . 1991
[7]  
Extraction of high-resolution frames from video sequences. Schultz R,Stevenson R. IEEE Transactions on Image Processing . 1996
[8]  
High-resolution image recovery from image-plane arrays, using convex projections. Stark H,Oskoui P. Journal of the Optical Society of America (A) . 1989
[9]  
Efficient generalized cross-valida-tion with applications to parametric image restoration and resolution enhancement. Nguyen N,Milanfar P,Golub G. IEEE Transactions on Image Processing . 2001
[10]  
Fundamental performance limits in image registration. Robinson,D.,Milanfar,P. IEEE Transactions on Image Processing . 2004