首页 > 论文 > 光电工程 > 38卷 > 4期(pp:87-94)

采用双树轮廓波及压缩传感的多聚焦图像融合

Multi-focus Image Fusion Using Dual-tree Contourlet and Compressed Sensing

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

为了拓展压缩传感(CS)的应用潜力,提出一种结合双树轮廓波(DT-Contourlet)及CS 的多聚焦图像融合方法。该方法首先采用DT-Contourlet 对源图像进行分解,提取图像的多尺度信息及方向信息,克服传统轮廓波变换不具平移不变性的缺点。接着,在DT-Contourlet 域,将分解系数看成包含稠密和稀疏两部分;对稠密成份,根据散焦的表现形式,采用邻域梯度作为清晰度指标,用选择法实现融合处理;对稀疏成份,则在CS 框架下,通过少数线性测量的融合,依据L1 范数最小化,采用两步迭代收缩的重构算法,得到融合结果。实验表明,该方法重构时收敛速度比正交匹配追踪法快,且融合结果无论在视觉质量及定量指标上都明显优于传统方法。

Abstract

In order to expand the capability of Compressed Sensing (CS), a fusion method for multi-focus image using Dual-tree Contourlet (DT-Contourlet) and CS is proposed. First, the source images are decomposed using DT-Contourlet for extracting multiscale and direction information while overcoming the limitation of traditional contourlet which is lack of shift invariance. Then, in DT-Contourlet domain, the decomposition coefficients are treated as containing two components, i.e., dense and sparse components. The dense components are fused using selection method by introducing neighborhood gradient as clarity index to indicate the characteristics of defocus. The sparse components are fused under the framework of CS via fussing a few linear measurements by solving the problem of L1 norm minimization which is based on a two-step iterative shrinkage/threshold reconstruction algorithm. The experiments demonstrate that the convergence rate of reconstruction is faster than that of orthogonal matching pursuit. Meanwhile, the proposed method provides more satisfactory fusion results in terms of visual quality and quantitative criterion.

中国激光微信矩阵
补充资料

中图分类号:TP751;TP391

所属栏目:图像与信号处理

基金项目:浙江省自然科学基金项目(Y1080778);浙江省公益性技术应用研究计划项目(2010C33104);教育部科学技术研究重点项目(209155)

收稿日期:2010-12-21

修改稿日期:2011-02-17

网络出版日期:0001-01-01

作者单位    点击查看

金炜:宁波大学 信息科学与工程学院,浙江 宁波 315211
符冉迪:宁波大学 信息科学与工程学院,浙江 宁波 315211
叶明:宁波大学 信息科学与工程学院,浙江 宁波 315211
励金祥:宁波大学 信息科学与工程学院,浙江 宁波 315211

联系人作者:金炜(jw1969@hotmail.com)

备注:金炜(1969-),男(汉族),浙江兰溪人。副教授,博士,主要从事数字图像处理的研究。

【1】Ma H,Jia C,Liu S. Multisource image fusion based on wavelet transform [J]. International Journal of Information Technology (S0218-7957),2005,11(7):81-91.

【2】De I,Chanda B. A simple and efficient algorithm for multifocus image fusion using morphological wavelets [J]. Signal Processing(S0165-1684),2006,86:924-936.

【3】Minh N D,Martin V. The contourlet transform:an efficient directinal multiresolution image representation [J]. IEEE Trans. on Image Processing(S1057-7149),2005,14(12):2091-2106.

【4】Yue L,Minh N D. A new contourlet transform with sharp frequency localization [C]// Proceedings of the IEEE International Conference on Image Processing,Atlanta,USA,Oct 8-11,2006:1629-1632.

【5】陈新武,龚俊斌,刘玮,等. 基于复轮廓波变换的图像消噪[J]. 光电工程,2009,36(10):111-115.
CHEN Xin-wu,GONG Jun-bin,LIU Wei, et al. Image Denoising Based on Complex Contourlet Transform [J]. Opto-Electronic Engineering,2009,36(10):111-115.

【6】Nguyen T T,Oraintara S. The shiftable complex directional pyramid:Theoretical aspects [J]. IEEE Trans. on Signal Processing (S1053-587X),2008,56(10):4651-4659.

【7】Selesnick I W,Baraniuk R G,Kingsbury N G. The dual-tree complex wavelet transform [J]. IEEE Signal Processing Magazine(S1053-5888),2005,22(6):123-151.

【8】Candes E,Wakin M B. An introduction to compressive sampling [J]. IEEE Signal Processing Magazine(S1053-5888),2008, 48(4):21-30.

【9】Provost F,Lesage F. The application of compressed sensing for photo-acoustic tomography [J]. IEEE Trans. on Med. Imaging (S0278-0062),2009,28(4):585-594.

【10】Donoho D L,Tsaig Y,Drori I,et al. Sparse solution of underdetermined linear equations by stagewise orthogonal matching pursuit[EB/OL]. (2006-12)[2010-5]. http://www.cs.tau.ac.il/~idrori/StOMP.pdf.

【11】Berinde R,Indyk P. Sparse recovery using sparse random matrices [EB/OL]. (2008-3)[2010-5]. http://people.csail.mit.edu/indyk/ report.pdf.

【12】Donoho D L. For most large underdetermined systems of linear equations,the minimal ell-1 norm near-solution approximates the sparsest near-solution [J]. Communications on Pure and Applied Mathematics(S0010-3640),2006,59(7):907-934.

【13】Daubechies I,Derise M. An interative thresholding algorithm for linear inverse problem with a sparsity constrain [J]. Communications on Pure and Applied Mathematics(S0010-3640),2004,57(11):1413-1457.

【14】Bioucas-dias J M,Figueirdo M A. A new TwIST: Two-Step Iterative Shrinkage/Thresholding algorithms for image restoration [J]. IEEE Trans. on Image processing(S1057-7149),2007,16(12):2992-3004.

【15】HAN Bing,WU Feng,WU Da-peng. Image representation by compressed sensing [C]// Proceedings of the International Conference on Image Processing,San Diego,USA,Oct 12-15,2008:1344-1347.

引用该论文

JIN Wei,FU Ran-di,YE Ming,LI Jin-xiang. Multi-focus Image Fusion Using Dual-tree Contourlet and Compressed Sensing[J]. Opto-Electronic Engineering, 2011, 38(4): 87-94

金炜,符冉迪,叶明,励金祥. 采用双树轮廓波及压缩传感的多聚焦图像融合[J]. 光电工程, 2011, 38(4): 87-94

被引情况

【1】陈广秋,高印寒,段锦,林杰. 基于LNSST与PCNN的红外与可见光图像融合. 光电工程, 2014, 41(10): 12-20

【2】陈广秋,高印寒,段锦,林杰. 基于LNSST与PCNN的红外与可见光图像融合. 光电工程, 2014, 41(10): 12-20

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF