基于小波变换的中医脉象信号特征提取与分析

利用小波变换所具有良好的时——频局部化的能力和对非平稳信号突变点的检测能力。实现对脉象信号同时进行时域、频域特征值的提取和分析；并提取了脉象在不同时间——尺度上的能量特征，可以作为脉象的新的特征值，用于脉象信号的辨识。经1456例临床脉象检测，不仅极大地提高对平、浮、沉、迟、数、虚、实、滑、涩、洪、弦、促、结、代等基本脉的识别率（准确率〉90％），对于由上述基本脉构成的临床常见的相兼脉也有相当高的识别能力（准确率〉82％）。

Pulse signal feature extraction and analysis based on wavelet transform

Wavelet transform is applied to synchronous extraction and analyses for time-domain feature values and frequency-domain ones of pulse signals for its advantages on time-frequency localization and saltation site detection of non-stationary signal. The energy feature of the pulse on different time scales is selected as the new feature value to distinguish the pulse. With this new feature value, the recognition rate for such basic pulses as fast pulse, slow pulse, feeble pulse, hollow pulse and astringent pulse is more than 90%, and the rate for concurrent pulses is more than 82%.