摘要
在利用小波变换进行图像融合的基础上,针对局部方差和高通滤波各自的优缺点,研究了一种新的融合规则的选择方法,提出了基于局部方差和高通滤波的小波变换图像融合算法。先对各源图像进行小波分解,然后采用局部方差准则融合源图像各分解层的高频信息,再针对参与融合的全色波段图像各分解层的低频信息进行高通滤波,用滤波后的低频细节信息叠加多光谱相应层的低频分量,最后通过小波逆变换得到融合图像。实验结果表明,该方法确实有效,相对单一的利用局部方差,提高了保持细节的能力;相对只用高通滤波,提高了保持光谱信息的能力。
In view of the characteristics of both the local deviation and high -pass filtering in image fusion, and based on using wavelet transform to do image fusion, a new method for selecting fusion rule was studied. An image fusion algorithm based on local deviation and high - pass filtering of wavelet transform was proposed. This method adopted different fusion rules for high - frequency and low frequency of image. After 2 - D wavelet decomposition, the high - frequency information in each decomposed - level was fused by means of local deviation rule while the low - frequency information in each level of wave band was taken with high - pass filtering, and then its details were added into the low - frequency component of muhi - spectral image with corresponding decomposed - level. Moreover, the merged image was reconstructed by an inverse wavelet transform. The simulation results show： compared with other single method, the new method presented is clearly better in preserving spectral information and improving spatial presentation.
出处
《计算机仿真》
CSCD
2008年第8期223-226,共4页
Computer Simulation
基金
国家自然科学基金资助项目(No.60604020)
关键词
小波变换
局部方差
高通滤波
图像融合
Wavelet transform
Local deviation
High- pass filtering
Image fusion
