提出了基于一类新的小波具有紧支撑、对称性和正交性、伸缩矩阵为(20 02)的非张量积小波的图像融合新方法.首先根据非张量积小波理论,提出了一种新的二维4通道4× 4对称滤波器组的设计方法,并用此方法设计出一组具有上述性质的非张量积小波4× 4滤波器组,利用此滤波器组对参加融合的图像进行滤波;然后对低频部分采用取均值、高频部分采用基于局部窗口能量取大的融合算法对滤波后的图像进行融合;最后重构.并采用熵、交叉熵、互信息、均方根误差和峰值信噪比等指标对该方法的融合性能进行了客观评价.对可见光图像与红外图像、远红外图像与近红外图像、航空图像和卫星图像、多聚焦图像等其它多类图像的融合实验结果表明本方法有较好的融合效果,可得到无边缘失真的融合结果图像,其融合性能比采用同样融合算法的张量积Haar小波的融合方法的融合性能好.
参考文献
[1] | Hui Li, Manjunath B S, Mitra K. Multi-sensor image fusion using the wavelet transform [J]. CVGIP:Graphical Models and Image Processing, 1995, 57 (3): 235-245. |
[2] | Charles K.Chui. An Introduction to Wavelets. American [M]. New York: Academic Press, 1995. chapter 5 |
[3] | Daubechies I. Ten Lecture on Wavelets [M]. Vermont: Capital City Press, 1992. chapter 8. |
[4] | Kovacevic J, Vetterli M. Non-separable multidimensional perfect reconstruction filter bank and wavelet bases for Rn [J]. IEEE Trans. on Information Theory, 1992, 38(2): 533-555. |
[5] | Liu Bin, Peng Jiaxiong. Image fusion method based on short support symmetric nonseparable wavelet [J].International Journal of Wavelets, Multiresolution and Information Processing, 2004, 2(1): 87-98. |
[6] | Antoine Ayache. Some methods for constructing non-separable orthonormal, compactly support wavelet [J].Letter to the Editor in Applied and Computational Harmonic Analysis, 2001, 10: 99-111. |
[7] | Lai M J, Roach D. Nonseparable symmetric wavelets with short support [C] // Proceedings of SPIE Conference on Wavelet Applications in Signal and Image Processing Ⅶ, 1999, 3813: 132-146. |
[8] | Chen Qiuhui, Micchelli C A, Peng Silong, et al. Multivariate filter banks having matrix factorizations [J]. Siam J. Matrix Anal. Appl., 2003, 25(2): 517-531. |
[9] | Liu Guixi, et al. A multiscale contrast pyramid based image fusion scheme and its performance evaluation [J].Acta Optica Sinica (光学学报), 2001, 21(11): 1336-1342 (in Chinese). |
[10] | Qu Quihong, Zhang Dali, Yan Pingfan. Information measure for performance of image fusion [J]. IEE Electron.Lett., 2002, 7(38): 313-315. |
上一张
下一张
上一张
下一张
计量
- 下载量()
- 访问量()
文章评分
- 您的评分:
-
10%
-
20%
-
30%
-
40%
-
50%