基于Hermite插值的小波模极大值重构滤波的肌电信号消噪方法.pdfVIP

第 31 卷第 4 期 电 子 与 信 息 学 报 Vol.31No.4 2009 年 4 月 Journal of Electronics Information Technology Apr. 2009 基于 Hermite 插值的小波模极大值重构滤波的肌电信号消噪方法 罗志增 沈寒霄 (杭州电子科技大学机器人研究所 杭州 310018) 摘 要：为了消除混杂在肌电信号中的噪声，该文提出了基于 Hermite 插值的小波模极大值重构滤波的肌电信号 消噪方法。该方法先对肌电信号进行小波分解；其次，根据小波系数的奇异性，利用信号与噪声模极大值在小波尺 度上的不同变化特性，分离出信号与噪声；再次，用 Hermite 插值法重构小波系数；最后从重构的小波系数恢复 成去噪后的信号。实验结果表明，Hermite 插值的小波模极大值重构能有效地去除噪声，提高信噪比，且保留了肌 电信号的细节信息，为肌电信号的特征提取和模式识别创造了良好的条件。 关键词：肌电信号；小波变换；模极大值；Hermite 插值 中图分类号：TP24 文献标识码：A 文章编号：1009-5896(2009)04-0857-04 Hermite Interpolation-Based Wavelet Transform Modulus Maxima Reconstruction Algorithm’s Application to EMG De-noising Luo Zhi-zeng Shen Han-xiao (Robotics Research Institute of Hangzhou Dianzi University, Hangzhou 310018, China) Abstract: In order to eliminate the noise mixed in EMG signal, a Hermite interpolation- based wavelet transform modulus maxima reconstruction algorithm is proposed. The obtained EMG signal is decomposed using the wavelet transform; the singularity and different properties of the signal and noise of the wavelet coefficient modulus under the different scales of wavelet transformation are used to separate the signal and noise; then the wavelet coefficients are reconstructed using Hermite interpolation; Finally, the de-noised signal is got by the wavelet reconstructed algorithm. Experiments show that, the method has good performance in removing noise, improving the signal-to-noise ratio and reserving the detailed information, which brings in favorable conditions to feature extraction and pattern recognition of EMG. Key words: EMG; Wavelet