基于曲波变换的图像融合研究

Study on Curvelet-based Image Fusion

近年来,曲波变换由于其独特性而受到研究人员的日益关注。曲波(Curvelet)作为一种新的多尺度分析方法比小波更加适合提取图像中的细节信息,具有较高的逼近精度和更好的稀疏表达性能。将曲波变换引入图像融合,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。使用曲波变换,进行图像融合,并同其它方法进行了比对,结果表明,该方法优于小波等融合方法。

The Curvelet transform has received more and more attention in recent years due to its unique characteristics.Curvelet is a new multiresolution analyses,which can represents edges better than wavelets and is well suited for extracting detailed information from an image. Curvelet transform provides near-ideal sparsity of representation of both smooth objects and of objects with edges.Image fusion based on Curvelet can extract the feature of original image , and provide more information for new image. Experiments carried out on a image show that the proposed curvelet method quantitatively outperforms state-of-the art image fusion methods.