摘要
针对传统小波去噪时图像边缘被破坏因而丢失有用细节信息的问题,提出了一种基于图像边缘检测的小波阈值去噪新方法,即先对边缘图像和非边缘图像进行小波分解,然后分别对其进行阈值处理,最后重构得到去噪图像.实验结果表明,这种方法与传统小波变换的全局阈值去噪方法相比,在去噪的同时有效地保留了图像边缘信息,图像信噪比有明显的提高.
In light of the problem of destroying the image＇s edge and losing the image＇s detail information in the traditional wavelet denoising method, a new wavelet threshold denoising method based on edge detection is presented. The image edge and non-edge character in wavelet coefficients are ascertained by decomposing the noised image and edge image, then the edge coefficients of wavelet （ high frequency） decomposed and non-edge coefficients are dealt with thresholds, finally, the denoised image is obtained by reconstructing. The experimental resuhs show that compared with the commonly-used wavelet threshold denoising method, this method can keep an image＇s edge from demaging and the ratio of signal to noise is improved obviously.
出处
《天津工业大学学报》
CAS
2007年第5期64-67,共4页
Journal of Tiangong University
关键词
小波变换
边缘检测
LOG算子
小波阈值去噪
wavelet transform
edge detection
LoG operator
wavelet threshold denoising
