基于小波变换和均值滤波的图像去噪方法

将小波变换和均值滤波相结合提出了一种有效的图像去噪方法,先将含噪图像进行小波分解,获得不同频带的子图像.将低频近似图像保持不变,对水平、垂直和对角三个方向高频细节图像根据其特性采用三种不同形状的模板进行均值滤波,最后将低频近似图像与三个均值滤波后高频细节图像合成得到去噪后的图像.实验结果表明,该方法在降低了图像噪声的同时又尽可能地保留图像的细节,其去噪效果优于单一小波阈值法和均值滤波法.

Image-denoising method based on wavelet transform and mean filtering

Combined with wavelet transform and mean filtering, an efficient image-denoising method was presented. Firstly, noised image was decomposed to various frequency sub-band images by wavelet transform. According to the respective characteristics, high frequency sub-band images at horizontal, vertical and angular directions were denoised by three different filter templates, without any changes to the low frequency approximation image. Denoised-image was obtained by composing the three high frequency detail images with mean filtering and low frequency approximation image. Experiment result shows that the noise of the image is removed effectively. At the same time, the detail of the image is kept well. The method has better denoising effect than single wavelet thresholding method and mean filtering method.