计算机应用 ›› 2015, Vol. 35 ›› Issue (6): 1749-1752.DOI: 10.11772/j.issn.1001-9081.2015.06.1749

• 虚拟现实与数字媒体 • 上一篇    下一篇

基于预测稀疏编码的快速单幅图像超分辨率重建

沈辉, 袁晓彤, 刘青山   

  1. 江苏省大数据分析技术重点实验室(南京信息工程大学), 南京 210044
  • 收稿日期:2014-12-29 修回日期:2015-03-22 发布日期:2015-06-12
  • 通讯作者: 沈辉(1990-),男,江苏如东人,硕士研究生,主要研究方向:稀疏表示、并行计算;shenhui2013@qq.com
  • 作者简介:袁晓彤(1980-),男,江苏南通人,教授,博士,CCF会员,主要研究方向:机器学习、计算机视觉;刘青山(1975-),男,安徽庐江人,教授,博士,主要研究方向:图像分析、视频分析、机器学习。
  • 基金资助:

    国家自然科学基金资助项目(61402232);江苏省自然科学基金资助项目(BK20141003)。

Fast super-resolution reconstruction for single image based on predictive sparse coding

SHEN Hui, YUAN Xiaotong, LIU Qingshan   

  1. Jiangsu Key Laboratory of Big Data Analysis Technology (Nanjing University of Information Science and Technology), Nanjing Jiangsu 210044, China
  • Received:2014-12-29 Revised:2015-03-22 Published:2015-06-12

摘要:

针对经典的基于稀疏编码的图像超分辨率算法在重建过程中运算量大、计算效率低的缺点,提出一种基于预测稀疏编码的单幅图像超分辨率重建算法。训练阶段,该算法在传统的稀疏编码误差函数基础上叠加编码预测误差项构造目标函数,并采用交替优化过程最小化该目标函数;测试阶段,仅需将输入的低分辨图像块和预先训练得到的低分辨率字典相乘就能预测出重建系数,从而避免了求解稀疏回归问题。实验结果表明,与经典的基于稀疏编码的单幅图像超分辨率算法相比,该算法能够在显著减少重建阶段运算时间的同时几乎完全保留超分辨率视觉效果。

关键词: 图像超分辨率, 预测稀疏编码, 字典学习, 交替优化

Abstract:

The classic super-resolution algorithm via sparse coding has high computational cost during the reconstruction phase. In view of the disadvantages, a predictive sparse coding-based single image super-resolution method was proposed. In the training phase, the proposed method imposed a code prediction error term to the traditional sparse coding error function, and used an alternating minimization procedure to minimize the resultant objective function. In the testing phase, the reconstruction coefficient could be estimated by simply multiplying the low-dimensional image patch with the low-dimensional dictionary, without any need to solve sparse regression problems. The experimental results demonstrate that, compared with the classic single image super-resolution algorithm via sparse coding, the proposed method is able to significantly reduce the reconstruction time while maintaining super-resolution visual effect.

Key words: image super-resolution, predictive sparse coding, dictionary learning, alternative optimization

中图分类号: