A FUSION ALGORITHM OF HIGH SPATIAL AND SPECTRAL RESOLUTION IMAGES 
BASED ON ICA 
GuoKun Zhang* 3 LeiGuang Wang b Hongyan Zhang* c 
a The Faculty of Tourism and Geographical Science Jilin Normal University, 1301 Haifeng Street,Siping, Jilin Province, 
China, 136000 
b State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, 
129 Luoyu Road, Wuhan, China, 430079 
c .College of Urban and Environmental, Northeast Normal University,5268 Renmin Street,Changchun, Jilin Province, 
China, 13 0024 
KEYWORDS: ICA Transform, image fusion, multiresolution analysis 
ABSTRACT: 
Independent component analysis (ICA) is a recently developed linear data analysis method. By using ICA method, the correlation 
and redundancy of multispectral images can be eliminated. In detail, our algorithm can be divided into the following steps (as shown 
in figure 1).Firstly, ICA transform is operated on MS imagery, and then, we get three new independent bands. Secondly, the discrete 
wavelet transform with linear phrase is used to PAN image and independent components. Then, the rule for combining the ICA 
coefficients with corresponding wavelet planes of panchromatic band is determined. Finally, inverse ICA is used to get the pan- 
sharpened image. Compared to other algorithms of RS imagery fusion, our method reduces the data redundancy among MS image 
bands and also preserves the spectral fidelity of the MS imagery as methods based on wavelet. Experiment result shows that our 
method can avoid the artifacts in the fused images and fusion result is not sensitive to wavelet decomposition levels. 
1. INTRODUCTION 
Due to the physical constraint of the spatial information sensors, 
there is a tradeoff between spatial and spectrum resolution of 
remote sensing (RS)image (Aiazzi et al., ; Choi). The goal 
fusing multispectral (MS) low-resolution remotely sensed 
images with a more highly resolved panchromatic (PAN) image 
is to obtain a high-resolution multispectral image which 
combines the spectral characteristic of the low-resolution data 
with the spatial resolution of the panchromatic image.(Choi, 
2006).A fused product used for visual analysis may provide 
better visual efficiency than the source image, the imagery can 
also improve the classification accuracy.(Aiazzi et al., ; Choi, ; 
Hill et al.) 
Literature has a large collection of fusion methods, which can 
be simply classified into several groups. One kind is methods 
based on color space transformation, including HIS, Lab, YUV 
and so on. One kind is base on the statistic methods, such as 
PCA, Brovey transformation etc. Another group of fusion 
algorithm is based on multi-resolution analysis (MRA) such as 
pyramid decomposition and wavelet transformation (Aiazzi et 
al.,; Wang et al.,2005). 
Nowadays, the wavelet-based scheme for the fusion of 
multispectral and panchromatic imagery has become quite 
popular due to its ability to preserve the spectral fidelity of the 
MS imagery while improving its spatial quality. But not all 
kinds of wavelet transform are available for fusion problem. 
Some shift variance of the transform can lead to artifacts in the 
fused images. In order to avoid this problem, a novel fusion 
algorithm combined Independent component analysis (ICA) 
and wavelet filters with linear phrase was proposed in this paper. 
The rest part is arranged as follows: in part two, the concept of 
ICA and a first algorithm are introduced. In the third part, a 
novel high frequency injection model in ICA domain is 
proposed. Finally, the experiment result is analyzed in the part 
4. 
2. ICA AND FAST ALGORITHMS 
Independent component analysis (ICA) is a statistical method 
for transforming an observed multidimensional random vector 
into components that are statistically as independent from each 
other as possible, which is proposed by Jutten and Herault in 
1991 (C and J, 1991; C and J,1999). The implication for feature 
extraction in remote sensing has been found in many 
works(Zhang et al.,2006).In the simplest way (Comon,1994), 
the ICA model can be described as follows: there are n 
unknown statistically independent components 
S^,S 2 ,S 3 ,S 4 ,-’-S n , and theirs linear combinations with m 
scalar variables V,, V 2 , V 3 • • • V m can be observed. That is: 
v¡ =зд +a, 2 s 2 +o (3 i 3 +-4a A =^ а 1,Л < 
7=1 ) 
i - 1,2,3••• m 
1 
'a:E-mail:ldxyzgk@163.com;phone(+86-0434)3291780 
X: Corresponding author: E-mail:zhy@nenu.edu.cn;phone+8613074334258