【作者】 吴太阳；

【导师】 蔡坤宝；

【作者基本信息】 重庆大学 ， 信号与信息处理， 2007， 硕士

【摘要】 中医独特的诊断方法及治病的疗效是有目共睹的。随着传感器技术和计算机处理技术的发展,人们开始致力于脉诊的客观化研究,希望用现代科学技术的方法和仪器,推进中医脉诊的现代化,这也是本文进行研究的目的。本论文着重对小波分析的基本概念和基本理论进行了详细的阐述,并探讨了其物理意义,在利用多分辨率分析脉象信号时,对算法进行了推导、验证和应用,且给出了多分辨率分析的矩阵表达方式,着重分析了小波系数和尺度系数的具体含义,为脉象信号的多分辨率分析奠定了坚实的基础。本论文还对神经网络的基本概念和基本理论进行了详细的阐述,突出探讨了概率神经网络的算法探讨和分析,为模式识别提高了扎实的理论依据。小波分析是一种在时域和频域均具有良好局域性的分析方法,尤其适用于非平稳信号的处理。本文应用小波分析的多分辨率分析算法分析了15例海洛因吸毒者和22例正常人脉象信号。通过提取小波系数和尺度系数,找出了海洛因吸毒者与正常人脉象信号之间的显著差异,初步提出了用于划分吸毒者和正常人的判据,根据该判据,22例正常人全被检测出来,而吸毒者B13被误检为正常人。本文还在对脉象信号进行多分辨率分析的基础上,利用概率神经网络优良的聚类效应,对37例脉象信号样本(15例海洛因吸毒者和22例正常人脉象信号)进行模式分类,结果把15例吸毒者的脉象信号识别出来,没有一个误判。更多还原

【Abstract】 Traditional Chinese medicine all along receives publicity for its unique diagnostic method and particularly curative effects. With the development of sensor and computer technology, people hope to apply modern technology to human pulse diagnosis to reveal the essence and features of pulse phenomena scientifically, which is the main research aspect in this paper.This paper deduced the theorems and formulas of the wavelet transform, and discussed the physics meaning of them, applied and proved them in the processing of the pulse signals. At the same time, multiresolution analysis in matrix form is given to get the clear idea of the wavelet coefficients and the scalar coefficients, which lays the foundation in the processing of the pulse signals.This paper also deduced the theorems and formulas of the neural networks, and gave especial research on algorithm of the probabilistic neural network, which is much helpful for the model recognition.Wavelet transform is a good analytical method both in the time and the frequency domains, especially applicable for non-stationary signal processing. In this paper we analyze pulse signals of 15 heroin addicts and 15 healthy persons using the multiresolution analysis of wavelet transform. By means of the wavelet coefficients and scalar coefficients of the wavelet transform, we found the significant difference between the heroin addicts and the healthy persons, a primary criterion for measuring off the heroin addicts and the healthy persons was obtained. Based on this criterion, the 15 healthy persons were identified and 1 heroin addicts were misjudged.After analyzing the pulse signals using the multiresolution method, this paper also uses the probabilistic neural network to identify the 30 pulse signals. Because of the good recognition behavior of the probabilistic neural network, the 15 pulse signals from the heroin addicts are well picked up, with excellent results to the end.更多还原