Multi-resolution Morphological Operators for Electrocardiogram Signal Analysis

Multi-resolution decomposition scheme has been considered as a powerful tool for signal processing. This paper proposes a multi-resolution morphological filter based on coupled wavelet, a morphological multi-resolution decomposition scheme. In addition, as an improvement of the multi-resolution morphological filter, a morphological lifting filter has been presented using the lifting scheme. Applying the proposed filters to electrocardiogram (ECG) signal analysis, which requires the removal of impulsive noise and the detection of characteristic waveforms, the results are very satisfactory.