A THRESHOLD DENOISING METHOD BASED ON EMD
A THRESHOLD DENOISING METHOD BASED ON EMD
复制标题
一种基于EMD的阈值去噪方法
DOI：
--
复制
发表时间：
2013
期刊：
影响因子：
--
通讯作者：
Li Liang
中科院分区：
文献类型：
--
作者：
J. Huang;Jian Xie;Li Feng;Li Liang
基于经验模式分解（EMD）基于经验模式分解的方法可以大致分为：IMF提取方法和IMF阈值方法。为了解决如何在提取方法中选择IMF的问题和所选IMF的处理，提出了基于EMD的阈值DeNoising方法。在这种方法中，提供了能源观点中IMF选择的标准，并且首先选择将IMF提升为标准的IMF，然后通过将所有未选择的IMF的平均能量与每个选定的IMF的能量进行比较，而单一选择的IMF为确认，并通过阈值剥去。最后，通过求和所有选定的IMF来获得DeNoed信号。当前方法将软阈值denoising方法与IMF选择结合在一起，与其他denoising方法相比，该方法的有效性和优越性得到了验证。结果为改善工程中的脱氧作用提供了支持。
The denoising method based on empirical mode decomposition (EMD) can be broadly divided into: IMF extraction method and IMF threshold approach. Aiming to the problems of how to select IMFs in extraction method and the processing of the selected IMFs, a threshold denoising method based on EMD is put forward. In this method, the standard of IMF selection in energy viewpoint is offered, and the IMFs upping to the standard are selected firstly, then, through comparing the average energy of all unselected IMFs with the energy of each selected IMF, the singular selected IMFs are confirmed, and denoised by threshold. Finally, the denoised signal is obtained by summing up all selected IMFs. The current method combines the soft threshold denoising method with the IMF selection together, compared with other denoising methods, the effectiveness and superiority of the method is validated. The result provides support for improving the denoising effect in engineering.