曲波(Curvelet)作为一种新的多尺度分析方法比小波更加适合分析二维图像中的曲线或直线状边缘特征,而且具有更高的逼近精度和更好的稀疏表达能力。将curvelet变换引入图像融合,能够更好地提取原始图像的特征,为融合图像提供更多的信息。第二代curvelet理论的提出也使得其理论更易理解和实现。因此,提出了一种基于第二代curvelet变换的图像融合方法,首先将图像进行curvelet变换,然后在相应尺度上利用融合规则将curvelet系数融合,最后进行重构得到融合结果。对多聚焦图像进行了实验,采用均方误差、偏差指数和相关系数对融合结果进行了客观评价,并与基于小波变换的融合进行了比较,实验结果表明该方法除分解2层时与小波性能相当,取其他分解层数时均获得更好的融合效果。

Curvelet, as a new multiscale analysis algorithm, is more appropriate for the analysis of the image edges such as curve and line characteristics than wavelet, and it has better approximation precision and sparsity description. When the curvelet transform is introduced to image fusion, the characteristics of original images are taken better and more information for fusion is obtained. The proposal of the second generation curvelet theory makes it under-stood and implemented more easily. Then the second generation curvelet transform based image fusion method is proposed. Firstly, the source images are decomposed using curvelet transform, then the curvelet coefficients are fused with the fusion regular in the corresponding scales, and finally the fused coefficients are reconstructed to obtain fusion results. Multi-focus images are taken as experimental data, mean square error, difference coefficient, correlation coefficient are used to evaluate the results, and comparison with results based on wavelet transform is also carried out . The results show that when decomposition level is 2, this method gets fusion result of similar quality with wavelet, but for other decomposition levels this method gets much better fusion results than wavelet.