一种改进的小波阈值去噪方法
An Improved Wavelet Threshold Denoising Method
摘要
在小波阈值去噪中,阈值的大小是一个重要的选择。为了克服传统的阈值在每个尺度上固定不变的缺陷,提出一种对传统的小波阈值去噪改进的方法,改进的阈值可以随着分解层数的变化而变化,在实际中可灵活应用。采用信噪比和均方误差作为评价去噪性能的参数,并与传统的阈值进行了比较,仿真结果表明:改进的小波阈值去噪方法其去噪效果优于传统固定阈值、Stein无偏似然估计阈值、启发式阈值和极大极小阈值的小波阈值去噪方法,具有较高的实用价值。
Abstract
In the wavelet threshold denoising, it is important to choose the threshold. In order to overcome the shortcomings of traditional fixed thresholds on each scale, a new wavelet threshold denoising method is proposed in this paper. The improved threshold can be changed with the number of decomposed layers, as it is shown flexibility in practice situation. In this paper, Signal Noise Ratio (SNR) and Mean Square Error(MSE)are used as the parameters to evaluate the denoising performance. Compared with the traditional threshold, the simulation results show that the denoising effect based on the improved wavelet threshold denoising method is better than the traditional wavelet denoising based on sqtwolog, rigrsure, heursure and minimaxi wavelet threshold. The improved wavelet threshold denoising method has a high practical value.
张振凤, 威欢, 谭博文. 一种改进的小波阈值去噪方法[J]. 光通信研究, 2018, 44(2): 75. ZHANG Zhen-feng, WEI Huan, TAN Bo-wen. An Improved Wavelet Threshold Denoising Method[J]. Study On Optical Communications, 2018, 44(2): 75.