江苏省科技期刊可视化阅读平台

导航条

江苏科技信息 ›› 2015, Vol. 32 ›› Issue (23): 52-55.doi: 10.3969/j.issn.1004-7530.2015.23.021

• 论文 • 上一篇    下一篇

改进OMP算法及其在图像超分辨率重建的应用

阮祥, 陈斌   

  1. 南京邮电大学理学院,江苏南京,210023
  • 出版日期:2015-08-15 发布日期:2015-08-15

Improved OMP Algorithm and Image Super Resolution Reconstruction

Ruan Xiang, Chen Bin   

  • Online:2015-08-15 Published:2015-08-15

摘要： 压缩感知（CS）理论是实现低维信号重建高维信号的一种有效方法。文章首先提出一种广义正交匹配（GOMP）算法，与其他算法相比，它能在较低的复杂度下有效提高稀疏重建率。然后将其应用到基于单幅的图像高分辨率重建方法中。实验结果表明，文章给出的方法在图像超分辨率重建上效果更好。

关键词: 压缩感知, 稀疏信号, 改进OMP算法, 超分辨率重建

Abstract: Compressive sensing is a prospective approach to effectively compress signals by sparse reconstruction:First,the paper proposes a generalized OMP, which compared with OMP,ROMP and CoSaMP,can dramatically improve the recovery rate of sparse signals with very low computation complexity.In addition ,a method for a reconstructing super-resolution image from a single low-resolution image is applied.The experiments results show that the proposed method can improve the performance of super-resolution reconstruction.