基于匹配子波变换的呼吸罗音检测与分类

Crackle Detection and Classification Based on Matched Wavelet Transform

提出一种基于匹配子波变换的罗音检测及分类方法。我们首先建立罗音信号的数学模型，然后基于这一模型设计子波变换的母函数。原始信号经过连续子波变换后，采用“软门限”进一步滤除噪声。根据每一尺度信号的能量，我们可以确定针对该信号的最佳尺度，从而可以检测罗音及对其进行分类。文中给出了实验方法及实验结果。

In this paper, we present a method for crackle detection which is based on ‘matched’ wavelet transform. We first modeled crackles as a mathematical function. Then we designed a matched mother wavelet based on this model. Applying a soft threshold to the results of the continuous wavelet transform to suppress noise further, we obtained the optimal scale Crackles were detected based on the envelope of the signal at optimal scale, and could be classified based on energy distribution with scale. The theory, methods and experimental results are given in detail.