一种基于小波-Contourlet的改进图像融合算法

An Improved Image Fusion Algorithm Based on Wavelet-Based Contourlet Transform

为了解决小波变换中的方向有限性问题,并消除Contourlet变换中拉普拉斯金字塔分解存在的信息冗余,提出一种基于小波-Contourlet的区域梯度选择平均图像融合算法.该算法对低频采用加权平均规则,对高频则计算区域梯度,并采用选择平均规则,选取空间频率、交叉熵与偏差等客观评价数据.实验结果表明,在相同的融合规则下,基于小波-Contourlet变换的融合算法能够比单一的小波变换或Contourlet变换获得更好的融合结果.

In order to solve the problem of directional finiteness in wavelet transform,and reduce the redundancy of Laplacian pyramid decomposing in contourlet transform,an image fusion algorithm for regional gradient selection mean is proposed,based on the wavelet-based contourlet transform（WBCT）.In this algorithm,weighted mean rules are employed for low frequence while for high frequence,the regional gradient is calculated with the use of selection average rules.Evaluation of the experimental results according to objective criteria including spatial frequence,cross entropy and deviation demonstrates that the fusion algorithm based on WBCT is better than the one based on wavelet transform or contourlet transform with the same fusion rules.