摘要
1引言
利用小波去除信号或图像中噪声的方法在过去十年间得到了广泛的关注.大体上可以分为三个阶段:最初是将含噪信号作正交小波变换.然后对其系数取阈值得到去噪后的信号[1].
Selecting threshold is the most important in threshold.based nonlinear filtering by wavelet transform. In this paper, a novel adaptive threshold is proposed by minimizing a Bayesian risk (It is adaptive to subband because it depends on data-driven estimates of the parameters). Combining this thresholding method with Wiener filting can result a new denoising method. Expermental results show that the proposed method indeed remove noise significantly and retaining most image edges. The results compare favorably with the reported results in the recent denoising literature.
出处
《计算机科学》
CSCD
北大核心
2003年第9期70-71,共2页
Computer Science
基金
陕西省自然科学基金(2000SL02)