心电图波形基线漂移的数学形态滤波在线实时处理

Baseline Drift of Electrocardiograph Dealt with Mathematical Morphology Filter in Real-time

研究数学形态学滤波器消除心电图波形的基线漂移。选择Maragos类型形态滤波器,得到平结构元为消除波形基线漂移最佳形态滤波形式,信号的衰减幅度由信号数字频率与结构元长度的比值唯一决定,由此归纳出平结构元长度大于或等于要滤波除去的分量的横向宽度的结论。借助Matlab软件进行算法模拟,并移植到DSP硬件平台上进行工程调试,实验表明,该方法运算简单,花费时间少于1 ms,能有效并实时地滤除基漂。

We have done a research on mathematical morphology filter in order to eliminate baseline drift in the wave of electrocardiograph（ECG）.Marago morphology filter type was chosen and flat structuring element filtering was found to be the best morphology filter modality for eliminating baseline drift,and the ratio of ＂signal numerical frequency to length of structuring element＂ determined the attenuation magnitude of the signal eliminated.We have come to the coclusion that the length of flat structuring element ought to be greater than or equal to the width of the signal component to be eliminated.The arithmetical algorithm was simulated with Matlab software,and was transplanted to DSP hardware platform.The result of experiment has shown that the arithmetic operation is simple and the method spends a short time（less than 1ms） in eliminating the baseline drift of ECG effectively and instantly.