Image Super-Resolution Reconstruction via Improved Dictionary Learning Based on Coupled Feature Space

doi:10.3969/j.issn.0372-2112.2016.05.025

您当前的位置：

首页 >

文章列表页 >

Image Super-Resolution Reconstruction via Improved Dictionary Learning Based on Coupled Feature Space

更新时间：2025-07-16

- Image Super-Resolution Reconstruction via Improved Dictionary Learning Based on Coupled Feature Space
- Acta Electronica Sinica Vol. 44, Issue 5, Pages: 1189-1195(2016)
- 作者机构：
  
  1. 合肥工业大学计算机与信息学院,安徽,合肥,230009
  2. ,陕西,西安,710086
  3. 三江学院电子信息工程学院,江苏,南京,210012
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China (No.61371156);Science and Technology Research and Development Project of Anhui Province (No.1401B042019)
- DOI：10.3969/j.issn.0372-2112.2016.05.025
  CLC： TP391
- Published Online：25 May 2016，
  
  Published：2016
- 稿件说明：
移动端阅览
Image Super-Resolution Reconstruction via Improved Dictionary Learning Based on Coupled Feature Space[J]. Acta Electronica Sinica, 2016, 44(5): 1189-1195.
DOI：

Image Super-Resolution Reconstruction via Improved Dictionary Learning Based on Coupled Feature Space[J]. Acta Electronica Sinica, 2016, 44(5): 1189-1195. DOI： 10.3969/j.issn.0372-2112.2016.05.025.

摘要

针对目前基于字典学习的图像超分辨率重建效果欠佳或字典训练时间过长的问题

本文提出了一种耦合特征空间下改进字典学习的图像超分辨率重建算法.该算法首先利用高斯混合模型聚类算法对训练图像块进行聚类

然后使用更改字典更新方式的改进KSVD字典学习算法来快速获得高、低分辨率特征空间下字典对和映射矩阵.重建时根据测试样本与各个类别的似然概率自适应地选择最匹配的字典对和映射矩阵进行高分辨率重建.最后利用图像非局部相似性

将其与迭代反向投影算法相结合对重建后的图像进行后处理获得最佳重建效果.实验结果表明了本文方法的有效性.

Abstract

Image super-resolution reconstruction via Improved Dictionary Learning based on Coupled Feature Space is studied in the paper

in order to solve the following problems:1 the dictionary training process is time-consuming

2 the results are not satisfactory in the existing algorithms.In the proposed algorithm

at first

the Gaussian mixture model clustering algorithm is employed to cluster the training image blocks

secondly

quickly obtain high and low resolution feature space of dictionary and mapping matrix by using dictionary updating based on improved KSVD dictionary learning algorithm

and then

the Super-Resolution image is reconstructed according to the likelihood probability of test samples

in which each category adaptively selected the most matching dictionary and mapping matrix for high-resolution reconstruction

finally

the non-local similarity and iterative back-projection are exploited to furtherly improve the quality of the reconstruction image.The experimental results demonstrate the validity of the proposed algorithm.

关键词

Keywords

references

Views

540

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Multimorphology Sparsity Regularized Image Super-Resolution

Image Super-Resolution Algorithms Based on Sparse Representation of Classified Image Patches

Scene Text Image Super-Resolution Reconstruction Based on Perceiving Multi-Domain Character Distance

A Spatiotemporal Fusion Algorithm of Remote Sensing Images Based on Cross-Scale Similarity Prior

Sparse ISAR Imaging Combined with Nearest Neighbor Graph Model

Related Author

SUN Yu-bao

WEI Zhi-hui

XIAO Liang

ZHANG Zhen-rong

LIAN Qiu-sheng

ZHANG Wei

HUANG Jun-yang

Related Institution

Lab of Pattern Recognition and Artificial Intelligence,Institute of Computer Science and Technology, Nanjing University of Science and Technology

Scientific Research Department of Military Training, 60th Research Institute of General Staff Department,Chinese People's Liberation Army

Lab of Pattern Recognition and Artificial IntelligenceInstitute of Computer Science and Technology Nanjing University of Science and TechnologyNanjingJiangsu 210094China

Scientific Research Department of Military Training 60th Research Institute of General Staff DepartmentChinese People's Liberation ArmyNanjingJiangsu 210016China

Institute of Information Science and Technology,Yanshan University

⁰