摘要
提出一种基于双窗口小波-Contourlet变换的图像融合算法。首先采用小波-Contourlet变换对图像进行分解,接着采用两种不同尺寸的窗口计算系数的特征值:低频子带采用区域能量法和高频子带采用区域方差法,分别比较双窗口下的高频子带和低频子带的系数特征值,比较的结果作为得到融合系数选择的依据,最后采用逆小波-Contourlet变换得到融合图像。仿真实验证实,在特征类型选择一致的情况下,该方法的融合性能优于小波变换法和Contourlet变换法,融合图像质量较好。
An algorithm for image fusion by wavelet-contourlet transform and double windows is proposed in this paper.Firstly,the image is decomposed by means of wavelet-contourlet transform.Secondly,the characteristic value of coefficients is calculated in two windows of different sizes：the value of low-frequency sub-band calculated through regional energy and the value of high-frequency sub-band calculated through regional variance.Meanwhile,the characteristic values under the two sub-bands in the double windows are compared respectively and the results of the comparison are taken as the selection basis of fusing the coefficients.Finally,a fused image is obtained through inversed wavelet-contourlet transform.Simulation experiments indicates that,in the same type of feature selection,the fusion effect of this method is better than that of wavelet transform and contourlet transform and leads to better fused images.
出处
《梧州学院学报》
2010年第3期1-9,共9页
Journal of Wuzhou University
基金
广西自然科学基金资助项目(0832283)
梧州学院基金资助项目(2008D011)
