基于相邻尺度积系数的半软阈值小波滤波
doi: 10.3724/SP.J.1146.2005.01699
Denoising by Multiscale Product Coefficient Semi-soft Thresholding
-
摘要: 针对相邻尺度积系数硬阈值滤波后的MSE函数不连续，最优阈值选取困难，该文构造了一种基于相邻尺度积系数的半软阈值函数。其为收缩因子函数与小波系数的乘积，对小波系数无穷可导、可在阈值邻域内对小波系数自适应收缩。进而通过极小化SURE(Stein Unbiased Risk Estimate)估计得到MSE意义下自适应于信号和噪声的最优阈值。大量仿真实验表明：采用本文构造的半软阈值函数，可改善基于相邻尺度积系数的滤波算法性能。Abstract: In multiscale product coefficient hard thresholding, how to determine the optimal threshold is the main problem due to the discontinuity of MSE. Here a semi-soft thresholding function is constructed in the product form of shrinkage coefficient function and wavelet coefficients. This function is infinite-order differentiable with respect to wavelet coefficient, and can adaptively shrink wavelet coefficient in the neighborhood of the threshold. Through minimizing the Stein Unbiased Risk Estimate (SURE) based on the function, the optimal threshold, varying with the signal and noise, is obtained in the Mean Square Error (MSE) sense. In simulations to denoise multiple classic noisy signals, the multiscale product coefficient thresholding is improved through our semi-soft thresholding function.
-
Donoho D L. De-noising by soft-thresholding [J].IEEE Trans. on Information Theory.1995, 41(3):613-627[2]Pan Quan, Zhang Lei, and Dai G Zh, et al.. Two denoising methods by wavelet transform [J].IEEE Trans. on Signal Processing.1999, 47(12):3401-3406[3]Gao Hong-Ye and Bruce A G. WaveShrink and semisoft shrinkage. StaSci Research Report No. 39, 1995.[4]Zhang X P and Desai M D. Thresholding neural network for adaptive noise reduction [J].IEEE Trans. on Neural Networks.2001, 12(3):567-584[5]Xu Y S, Weaver J B, and Healy D M, et al.. Wavelet transform domain filters: a spatially selective noise filtration technique [J].IEEE Trans. on Image Processing.1994, 3(6):747-758[6]Zhang Lei and Bao Paul. Denoising by spatial correlation thresholding [J].IEEE Trans. on Circuits and System for Video Technology.2003, 13(6):535-538[7]Jansen M. Noise reduction by wavelet thresholding. Springer Verlag, Lecture notes in Statistics, vol. 161: 2001.[8]潘泉，戴冠中，张洪才. 具有理论阈值的小波域滤波方法[J]. 宇航学报, 1998, 19(4): 81~85.Pan Quan, Dai Guan-zhong, and Zhang Hong-cai. A threshold selection method for wavelet transform domain filter. Journal of Astronautics, 1998, 19(4): 81-85.[9]Mallat S and Zhong S. Characterization of signals from multiscale edges [J].IEEE Trans. on Pattern Analysis and Machine Intelligence.1992, 14(7):710-732[10]Donoho D L and Johnstone I M. Adapting to unknown smoothness via wavelet shrinkage [J].Journal of the American Statistical Association.1995, 90(432):1200-1224[11]Stein C. Estimation of the mean of a multivariate normal distribution[J].The Annals of Statistics.1981, 9(6):1135- -
计量
- 文章访问数: 3468
- HTML全文浏览量: 133
- PDF下载量: 889
- 被引次数: 0
下载:
