采用改进投影梯度非负矩阵分解和非采样Contourlet变换的图像融合方法

杨粤涛; 朱明; 贺柏根; 高文

doi:10.3788/OPE.20111905.1143

您当前的位置：

首页 >

文章列表页 >

采用改进投影梯度非负矩阵分解和非采样Contourlet变换的图像融合方法

信息科学 | 更新时间：2020-08-12

- 采用改进投影梯度非负矩阵分解和非采样Contourlet变换的图像融合方法
- Fusion algorithm based on improved projected gradient NMF and NSCT
- 光学精密工程 2011年19卷第5期页码：1143-1150
- 作者机构：
  
  1. 中国科学院长春光学精密机械与物理研究所,吉林长春 130033
  2. 中国科学院研究生院,北京 100039
- 作者简介：
- 基金信息：
  
  国家863高技术研究发展计划资助项目(No.2005AA778032)()
- DOI：10.3788/OPE.20111905.1143
  中图分类号： TP391.4
- 收稿日期：2010-08-04，
  
  修回日期：2010-09-13，
  
  网络出版日期：2011-05-26，
  
  纸质出版日期：2011-05-26
- 稿件说明：
移动端阅览
杨粤涛, 朱明, 贺柏根, 高文. 采用改进投影梯度非负矩阵分解和非采样Contourlet变换的图像融合方法[J]. 光学精密工程, 2011,19(5): 1143-1150

YANG Yue-tao, ZHU Ming, HE Bai-gen, GAO Wen. Fusion algorithm based on improved projected gradient NMF and NSCT[J]. Editorial Office of Optics and Precision Engineering, 2011,19(5): 1143-1150
杨粤涛, 朱明, 贺柏根, 高文. 采用改进投影梯度非负矩阵分解和非采样Contourlet变换的图像融合方法[J]. 光学精密工程, 2011,19(5): 1143-1150 DOI： 10.3788/OPE.20111905.1143.

YANG Yue-tao, ZHU Ming, HE Bai-gen, GAO Wen. Fusion algorithm based on improved projected gradient NMF and NSCT[J]. Editorial Office of Optics and Precision Engineering, 2011,19(5): 1143-1150 DOI： 10.3788/OPE.20111905.1143.

摘要

针对非负矩阵分解(NMF)算法时间复杂度较高

而投影梯度(Projected Gradient

PG)优化方法可以大幅降低NMF约束优化迭代问题的时间复杂度

提出一种基于改进的投影梯度NMF(IPGNMF)和非采样Contourlet变换(NSCT)相结合的图像融合方法。采用NSCT对已配准的源图像进行多尺度、多方向的分解

将分解后的低频部分作为原始数据

利用IPGNMF得到包含特征基的低通子带系数;高频部分应用了一种基于邻域一致性测度(NHM)的局部自适应融合规则得到各带通方向子带系数。经过NSCT逆变换得到融合图像。实验结果表明

融合结果在主观和客观评价上均优于NSWT方法、IPGNMF方法和NSCT方法。与NSCT法相比

实验所采用的两组图像的信息熵、清晰度和

指标分别提高了0.062 7%、0.901%、3.120 1%和 2.769%、2.203%、1.049%。

Abstract

As the Non-negative Matrix Factorization (NMF) algorithm has a higher iteration time complexity and the Gradient Projection(BP) optimization method can significantly reduce the NMF iteration time complexity

an image fusion algorithm by combing the Improved PGNMF(IPGNMF) and Nonsubsampled Contourlet Transform (NSCT) is proposed in this paper. Firstly

the registered original images are in multi-scale and multi-direction decomposition in NSCT domain. According to the characters of the different areas

different fusion rules are designed in the NSCT domain. The low-pass sub band coefficients used as original data impose to the IPGNMF algorithm to obtain the fused low-pass sub band coefficients and the band-pass directional sub band coefficients impose to the Neighborhood Homogeneous Measurement (NHM) algorithm to obtain the fusion band-pass directional sub band coefficients. Finally

the fused result is obtained through inverse NSCT. The proposed algorithm has been experimented on two groups of different scene images

and experimental results show that it superior to those conventional fusion methods based on NSWT

IPGNMF and NSCT in subjective and objective standards.As contrasted with NSCT method in two group images

its entropy

definition and

ABIF

have been increased by 0.0627%

0.901%

3.1201% and 2.769%

2.203%

1.049%

respectively。

关键词

Keywords

references

CUNHA A L, ZHOU J P, DO M N. The nonsubsampled Contourlet transform: theory,design and applications[J]. IEEE Trans on Image Processing, 2006,15(10):3089-3101.[2] LEE D D,SEUNG H S. Learning the parts of objects by non-negative matrix factorization [J]. Nature, 1999,401(6755):788-791.[3] LEE D D, SEUNG H S. Algorithms for non-negative matrix factorization . Advances in Neural Information Processing Systems 13, Denver, 2000:556-562.[4] 苗启广,王宝树. 基于非负矩阵分解的多聚焦图像融合研究[J]. 光学学报,2005,25(6):755-759. MIAO Q G,WANG B SH. Multi-focus image fusion based on Non-Negative Matrix Factorization [J].Acta Optica Sinica,2005,25(6):755-759.(in Chinese)[5] NOVAK M,MAMMONE R.Use of non-negative matrix factorization for language model adaptation in a lecture transcription task . Proc. IEEE International Conference on Acoustics, Speech and Signal Processing, SaltLake, 2001:541-544.[6] GUILLAMET D, BRESSAN M, VITRIA J. A weighted non-negative matrix factorization for local representations . Proc IEEE Computer Society Conference on Computer Vision and Pattern Recognition V1, Kauai, HI, 2001:942-947.[7] FENG T,LI S Z,SHUM H Y,et al.. Local non-negative matrix factorization as a visual representation . Proc.2nd International Conference on Development and Learning,Cambridge, 2002:1-6.[8] DO M N, VETTERLI M. The Contourlet Transform: an efficient directional multiresolution image representation [J]. IEEE Trans . on Image Processing , 2005,14(12):2091-2106.[9] BAMBERGER R H, SMITH M T. A filter bank for the directional decomposition of images: theory and design [J].IEEE Trans. on Signal Processing, 1992,40(4):882-893.[10] LEE D D,SEUNG H S. Algorithms for Non-negative Matrix Factorization [J]. Advances in Neural Information Processing, MIT Press, 2001,13:556 -562.[11] 李乐,章毓晋. 非负矩阵算法综述[J]. 电子学报,2008,36(4):737-743. LI L, ZHANG Y J. A survey on algorithms of Non-Negative Matrix Factorization[J]. Acta Electronica Sinica, 2008,36(4):737-743.(in Chinese)[12] LIN C J. Projected gradient methods for non-negative matrix factorization [J].Neural Computation,2007,19(10):2756-2779. [13] LANGGREBE D A. Multispectral Data Analysis: A Signal Theory Perspective . http://cobweb.ecn. purdue.edu/~biehl/MultiSpec/documentation.htm1[14] 胡良梅,高隽,何柯峰. 图像融合质量评价方法的研究[J]. 电子学报,2004,32(12A):218-221. HU L M,GAO J,HE K F. Research on Quality Measures for Image Fusion[J]. Acta Electronica Sinica, 2004,32(12A):218-221.(in Chinese)[15] 屈小波,闫敬文,杨贵德. 改进拉普拉斯能量和的尖锐频率局部化Contourlet域多聚焦图像融合方法[J]. 光学精密工程,2009,17(5):1203-1211. QU X B,YAN J W,YANG G D.Multifocus image fusion method of sharp frequency localized Contourlet transform domain based on sum-modified-Laplacian[J]. Opt. Precision Eng., 2009,17(5):1203-1211.(in Chinese)

浏览量

610

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

采用非采样Contourlet变换与区域分类的红外和可见光图像融合

电力巡检中的偏振图像特征融合

基于多尺度空间注意力互补的红外与可见光图像融合

自适应色彩补偿和多尺度融合的水下图像增强

基于直觉模糊集熵测度和显著特征检测的古铜镜X光图像融合