高级检索

基于非子采样Contourlet变换的图像融合方法

Image Fusion Method Based on the Nonsubsampled Contourlet Transform

  • 摘要: 分析了非子采样Contourlet变换滤波器组的设计与实现方法,提出一种基于非子采样Contourlet变换的图像融合方法.首先将图像作非子采样拉普拉斯金字塔尺度分解,并在各尺度层使用非子采样方向滤波器组对高频子带作方向分解,构成非子采样Contourlet变换;然后,采用基于区域能量的融合规则得到融合图像的非子采样Contourlet系数;最后进行非子采样Contourlet逆变换得到融合图像.实验结果表明,该方法的融合效果优于à trous小波变换方法和Mallat小波变换方法.

     

    Abstract: An image fusion algorithm based on the nonsubsampled Contourlet transform (NSCT) is proposed by analyzing the design and implementation of the filter bands of the NSCT.By the algorithm,firstly,multiscale decomposition of the source images is performed with nonsubsampled Laplacian pyramid.Directional decomposition of the high-frequency subbands in every scale is also made with nonsubsampled directional filter bands.Multiscale decomposition and directional decomposition constitute the NSCT.And then, the NSCT coefficients for the fused image can be obtained by means of the fusion rule based on region energy.Finally,the fused image is reconstructed by inverse NSCT.Experimental results show that the method presented in the paper works better than the à trous wavelet transform and the Mallat wavelet transform.

     

/

返回文章
返回