基于耦合小波阈值的图像放大增强方法

利用单尺度平移不变Haar小波变换和扩散方程离散一步迭代的等价性,提出一种图像放大增强的新方法。该方法把原图像作为放大图像的小波低频子带,对重构后的图像进行平移不变小波分解,然后进行基于前向-后向扩散方程的耦合小波阈值。该阈值使得在放大的同时能更好地去除噪声和增强图像的边缘部分,放大和增强同时实现。该方法简单易实现,数值实验结果证实,此方法为一个有效的图像放大增强法。

Image zooming and enhancing based on coupled wavelet shrinkage

A new image zooming and enhancing method is proposed using the equivalence between single-level shift invariant Haar wavelet transform and the discrete one iteration of diffusion equations.The original image is regarded as the low frequency of the wavelet analysis of the zoomed one,and then the shift invariant wavelet analysis is applied with the forward-backward diffusion-based coupled wavelet shrinkages.The image is zoomed and denoised as well as edge enhancement with this type of shrinkage.The proposed method is easy to implement and some numerical experiments illustrate that it is a valid new image zooming and enhancing algorithm.