摘要
文章提出了基于小波变换能量加权的噪声滤除方法。先将含噪图像进行小波分解,获得低频近似图像和水平、垂直、对角三个高频细节图像,然后根据其特性进行能量加权,最后将能量加权处理后的低频近似图像和三个高频细节图像合成得到去噪后的图像。实验结果表明,与传统小波变换和中值滤波、均值滤波相比,该方法峰值信噪比提高了约0.7dB,在降低了图像噪声的同时又尽可能地保留了图像的细节信息。
This paper puts forward an effective image-denoising method using wavelet transform combined with weighted energy. Firstly the noised image should be decomposed by wavelet transform to get low-frequency approximation image and three high-frequency detailed images, which are horizontal, vertical and diagonal ones. And then the images are weighted energy according to their characteristics. Finally the de-noising image is obtained by synthesizing the energy weighted low-frequency approximation image and three high-frequency detailed images. The experimental results show that the proposed method improves the peak signal to noise ratio by about 0.7 dB compared with the traditional wavelet transform, the median filtering and the average filtering methods. It can reduce the noise of images while retaining details of the images as far as possible..
出处
《电子技术(上海)》
2013年第6期16-19,共4页
Electronic Technology
基金
曲阜师范大学2012年大学生科学技术立项项目资助
关键词
图像去噪
小波变换
能量加权
中值滤波
均值滤波
image denoising
wavelet transform
energy weighting
median filtering
mean filtering