期刊文献+

基于Harr小波的CS-MRI典型重构算法的性能分析

Performance Analysis of CS-MRI Recovery Algorithms Based on Harr Wavelet
下载PDF
导出
摘要 目的压缩感知理论(Compressed Sensing,CS)与磁共振成像(Magnetic Resonance Imaging,MRI)相结合,缩短MRI图像数据的扫描时间,提高成像质量。方法以Harr小波进行稀疏表达,分别利用基追踪(Basis Pursuit,BP)算法、正交匹配追踪(Orthogonal Matching Pursuit,OMP)算法和分段正交匹配追踪(Stagewise Orthogonal Matching Pursuit,St OMP)算法实现CS-MRI的二维重构。结果在采样率较低(10%-50%)时,以峰值信噪比(Peak Signal to Noise Ratio,PSNR)、平均结构相似度(Mean Structure Similarity,MSSIM)、相对误差(Relative L2 Norm Error,RLNE)和传输边缘信息(Transferred Edge Information,TEI)四个指标来定性、定量地评价和比较上述三种算法的重构质量,BP算法性能最佳。结论 BP算法能精确重构原始图像,与完整采样图像相比,图像质量并无明显下降,同时大大减少MRI采集时间,具有重要的理论意义和临床应用价值。 Objective The CS(Compressed Sensing) theory is introduced into MRI(magnetic resonance imaging) to solve the shortcomings of the long acquisition time of MRI, and to change the way of medical imaging techniques. Methods Three typical CS-MRI recovery algorithms, including basis pursuit(BP), orthogonal matching pursuit(OMP) and stagewise orthogonal matching pursuit(St OMP) are analyzed and compared based on the Harr wavelet. Results Four performance indices-peak signal to noise ratio(PSNR), mean structural similarity(MSSIM), relative l2 norm error(RLNE) and transferred edge information(TEI) are applied to evaluate the recovery image's quality under different sampling ratios. It is clear that the BP algorithm achieves the best performance. Conclusion The original reference images can be accurately recovered based on BP algorithm under the low sampling ratios(10%-50%), which has important theoretical significance and clinical application foreground in MRI technique.
作者 任筱倩 汤敏
出处 《智慧健康》 2016年第5期14-22,共9页 Smart Healthcare
基金 国家自然科学基金资助项目(61401239) 江苏省自然科学基金资助项目(BK20130393) 江苏高校品牌专业建设工程资助项目(PPZY2015B135)
关键词 压缩感知 重构算法 小波变换 Harr小波 MRI图像 Compressed sensing Reconstruction algorithm Wavelet transform Harr wavelet MRI images
  • 相关文献

参考文献3

二级参考文献44

  • 1Deanna Needell,Roman Vershynin.Uniform Uncertainty Principle and Signal Recovery via Regularized Orthogonal Matching Pursuit[J]. Foundations of Computational Mathematics . 2009 (3)
  • 2Chen S,Donoho DL,Saunders MA.Atomic decomposition by basis pursuit. SIAM Journal on Scientific Computing . 1998
  • 3Neff R,Zahor A.Very low bit rate video coding based on matching pursuits. IEEE Transactions on Circuits and Systems for Video Technology . 1997
  • 4Donoho D L.Compressed sensing. IEEE Transactions on Information Theory . 2006
  • 5Richard G Baraniuk,Emmanuel Candes,Robert Nowark,Martin Vetterli.Compressive sampling. Signal Processing . 2008
  • 6Tropp J,Gilbert A.Signal recovery from random measurements via orthogonal matching pursuit. Transactions on Information Theory . 2007
  • 7D. Needell,J. A. Tropp."CoSaMP: Iterative Signal Recovery from Incomplete and Inaccurate Samples". ACM Technical Report 2008 . 2008
  • 8DAI W,MILENKOVIC O.Subspace pursuit for compressive sensingsignal reconstruction. IEEE Transactions on Information Theory . 2009
  • 9CAND S E,TAO T.Decoding by linear programming. IEEE Transactions on Information Theory . 2005
  • 10Candes E.The restricted isometry property and itsimplications for compressed sensing. ComptesRendus Mathematique . 2008

共引文献100

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部