Pulse wave signal feature point detection method based on waveform time domain features

The invention provides a pulse wave signal feature point detection method based on waveform time domain features. The mode that the interval of analyzing waveform is gradually decreased is adopted, the positions of feature points in pulse wave signals are determined by combining a differential algorithm with a wavelet transform algorithm on the basis of the waveform time domain features of the pulse wave signals in the analyzing process, therefore, unnecessary interference can be effectively removed, the signal features are amplified, the accuracy of pulse wave signal feature point detection is improved, and the recognition accuracy for the condition that tidal waves or dicrotic waves in pulse wave waveforms are not obvious is high. According to the pulse wave signal feature point detection method based on the waveform time domain features, detection of a computer on the pulse wave signal feature points is achieved, the workload of clinicians is reduced, the defect of manual detection errors is overcome, and a technological base is supplied to automatic calculation and acquisition of clinical pulse wave information by computer equipment and research and development of continuous non-invasive blood pressure detection equipment.