利用双正交小波变换去除图象噪声

对噪声污染的图象进行复原处理时 ,传统的去噪声方法会造成图象模糊。本文用双正交小波变换去除图象噪声 ,从结果看利用双正交小波变换进行多分辨率低通滤波在去除噪声的同时造成的模糊比传统低通滤波器少

Biorthogonal Wavelet Transforms for Removing Noise in Digital Images

Images obtained from an image are always degraded for several reasons and there are generally ill posed problems when they are restored. In this paper we try to remove the noise using biorthogonal wavelet transforms. By using chosen scale functions and wavelet functions, the image of resolution target is decomposed into biorthogonal multiresolution representation. We chose a threshold value to remove the noise in different resolutions. Then, the decomposed images having been processed are reconstructed. We show that by using multiresolution low past filter, based on biorthogonal wavelet transforms, higher frequency information of the images is kept than in the case of traditional method while the noise is reduced. For comparing purpose we also give the result of median filter.