论文部分内容阅读
本文介绍了两种对超声多普勒血流声谱国进行波形自动分类决策的方法。方法一，与统计得到的标准波形组对比，按适当判推进行自动分类；方法二，送入经训练的人工神经网络进行自动分类。待分类波形由计算多普勒信号的最大频率提取声谱包络并经分段、平均和归一化等处理得到。文中给出了计算机模拟的结果。
This article describes two methods for automatic waveform classification of ultrasound Doppler sonography. The first method is to compare with the standard waveform sets obtained by statistics and classify them according to the proper judgment. Method two is to send into the trained artificial neural network for automatic classification. The waveform to be classified is extracted from the maximum frequency of the Doppler signal to extract the spectral envelope and is processed by segmentation, averaging and normalization. The paper gives the result of computer simulation.