Computer Engineering and Applications ›› 2014, Vol. 50 ›› Issue (24): 209-212.
Previous Articles Next Articles
GUO Xingming, HE Yanqing, LU Delin, YUAN Zhihui
Online:
Published:
郭兴明,何彦青,卢德林,袁志会
Abstract: The wavelet threshold de-noising method can eliminate the noise existing in heart sound signals, but the lack of translation invariance, may produce artificial oscillation phenomenon near the singularity of the signal, namely Pesudo-Gibbs phenomenon influencing the de-noising result. Therefore, the wavelet threshold de-noising method based on translation invariance(Translation Invariance, TI) is adopted to process the heart sound signals, which changes the position of singular points through translating the signal sequence, to reduce or eliminate the oscillation. Before processing heart sounds with this way, firstly it needs to eliminate the interference introduced during the acquisition process by reducing trend terms. The experimental results show that this method eliminates artificial oscillation phenomenon, and that the Signal-to-Noise Ratio(SNR) and Root Mean Square Error(RMSE) are obviously improved under the precondition of keeping the main characters of the heart sound signals.
Key words: heart sound signals, wavelet threshold de-noising, translation invariance
摘要: 小波阈值去噪方法可以消除心音信号中的噪声,但其缺乏平移不变性,可能在信号的奇异点附近产生人为的振荡现象,即Pesudo-Gibbs现象,影响去噪效果。采用平移不变(Translation Invariance,TI)小波阈值去噪的方法对心音信号进行去噪,通过对信号序列平移来改变奇异点在整段信号的位置,以降低或消除振荡。对信号采用平移不变小波去噪之前,先通过消除趋势项来降低信号采集过程中引入的干扰。实验结果表明,该方法消除了人为振荡现象,在保留心音信号主要特征的前提下,信号的信噪比(Signal-to-Noise Ratio,SNR)和根均方误差(Root Mean Square Error,RMSE)均得到明显改善。
关键词: 心音信号, 小波阈值去噪, 平移不变性
GUO Xingming, HE Yanqing, LU Delin, YUAN Zhihui. Application of translation invariant wavelet in heart sound signal de-noising[J]. Computer Engineering and Applications, 2014, 50(24): 209-212.
郭兴明,何彦青,卢德林,袁志会. 平移不变小波在心音信号去噪中的应用[J]. 计算机工程与应用, 2014, 50(24): 209-212.
0 / Recommend
Add to citation manager EndNote|Ris|BibTeX
URL: http://cea.ceaj.org/EN/
http://cea.ceaj.org/EN/Y2014/V50/I24/209