摘要
采用均值滤波和小波变换相结合的图像去噪方法是先将含噪图像进行小波分解,在小波域中,选取适当的阈值对小波系数进行处理,然后对图像信号进行局部重构至第一层,并采用不同的模板对第一层的各细节子图像进行均值滤波,最后将低频近似图像与均值滤波后高频细节图像合成得到去噪后的图像。这种方法与全局Donoho软、硬阈值小波去噪方法和Birge-Massart策略软、硬阈值小波去噪方法相比,其去噪效果更为明显。它在降低了图像的噪声的同时,又尽可能地保留图像的细节,且图像更加平滑。仿真实验证明,该方法是一种有效的图像去噪方法。
A method of image de-noising technology based on the mean filtering and wavelet transformation is proposed.Firstly noised-image was decomposed by wavelet transformation and the wavelet parameters were processed using reasonable threshold value in the domain of wavelet.Then the image can be reconstructed correctly to the first layer and according to respective characteristics every sub-band images were denoised by different filter templates.Lastly denoised-image was obtained by composing the three high frequency detail images with mean filtering and low frequency approximation image.Compared with the wavelet de-noising methods of global Donoho soft,hard threshold and the wavelet de-noising methods of Birge-Massart strategy soft,hard threshold,the de-noising effect of this way is more obvious.It not only reduces the noise of the image,but also retains the detail of the image,and the image looks more smoother.The experiment proves this method is an effect way of de-noising image.
出处
《计算机技术与发展》
2011年第2期51-53,57,共4页
Computer Technology and Development
基金
山东省教育科技计划项目(J08LJ64)
关键词
图像噪声
均值滤波
小波变换
去噪
Image noise
mean filtering
wavelet transformation
de-noising
