基于单尺度脊波变换的图像融合

图像融合在一定程度上可以理解为对多个二维函数的奇异信息进行分析、提取、综合的过程。单尺度脊波变换的函数逼近性能要优于小波变换，因此，对应于图像中的边缘及角点的单尺度脊波系数的能量要更加集中。分析了基于变换域的图像融合算法的性能与所用变换的函数逼近性能的关系，提出了一种新的基于单尺度脊波变换的图像融合算法。在多种融合规则下，将该方法与基于Laplacian塔型变换、小波变换等其他图像融合方法进行了比较，实验结果表明，基于单尺度脊波变换的融合方法具有更好的融合效果。

Image fusion based on monoscale ridgelet transform

Image fusion can be considered as a process to analyze, abstract and synthesize singularities of 2-D functions. The upper bound of approximation error of the monoscale ridgelet transform is lower than that of the wavelet transform. As a result, the energy of the coefficients correlated with the edges and the comers in images is more compact. The relation between the effect of fusion methods based on the transform domain and the rate of approximation of the transform was discussed. A new image fusion method based on the monoscale ridgelet transform was put forward. In experiments, the proposed method was compared with the algorithms based on the Laplacian pyramid, the wavelet transform, and other existing methods. Results show that the proposed method outperforms the others, and exploit the application scope of the monoscale ridgelet transform.