首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于心音窗函数的心音图形化处理方法的研究
引用本文:成谢锋,李伟.基于心音窗函数的心音图形化处理方法的研究[J].物理学报,2015,64(5):58703-058703.
作者姓名:成谢锋  李伟
作者单位:南京邮电大学电子科学与工程学院, 南京 210003
基金项目:国家自然科学基金(批准号: 61271334)资助的课题.
摘    要:心音分析与识别目前主要局限在一维信号处理方面, 为了获得心音信号更直观特征表现形式, 提高分类识别效果, 拓展心音识别研究领域, 提出了一种将心音与图像处理技术相结合、基于心音窗函数的心音纹理图特征提取与识别算法. 本文首先给出心音的模型, 定义心音时频图和心音纹理图, 然后讨论如何利用心音窗函数和短时傅里叶变换获取二维心音时频图, 并且针对心音的特点, 重点研究了心音窗函数的构造原则和实现方法, 最后通过改进的脉冲耦合神经网络模型实现了对心音纹理图的特征提取与身份识别. 仿真实验表明, 心音窗函数与传统窗函数相比较, 所获得的心音时频图具有第一、第二心音纹理更加清晰, 噪声纹理得到较好抑制的优点, 并且改进的脉冲耦合神经网络模型具有更低的计算成本, 与3种典型识别方法相比较, 呈现更高的识别率, 因而基于图像处理技术对心音进行特征提取与身份识别是一种行之有效的方法.

关 键 词:心音时频图  心音纹理图  心音窗函数  改进的脉冲耦合神经网络
收稿时间:2014-06-27

Research on heart-sound graphical processing methods based on heart-sounds window function
Cheng Xie-Feng,Li Wei.Research on heart-sound graphical processing methods based on heart-sounds window function[J].Acta Physica Sinica,2015,64(5):58703-058703.
Authors:Cheng Xie-Feng  Li Wei
Institution:College of Electronic Science and Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210003, China
Abstract:Currently, the one-dimensional signal processing method for heart-sound analysis and recognition is the mainstream in researches. In order to gain more intuitive features in manifestation, to improve the effect of classification, and to endarge the heart-sound recognition field, this paper puts forward a heart-sound texture feature extraction and recognition algoriithm, which is based on heart-sound window function and the combination of heart-sound and image processing technology. Firstly, we give a heart-sound model, a definition of heart-sound time-frequency diagram, and a heart-sound texture map; we also discuss how to utilize heart-sound window function and short-time Fourier transform to obtain a two-dimensional heart-sound time-frequency diagram. After that, in the light of the characteristics of heart-sound, we mainly study the structure principle and implementation method of a heart-sound window function Finally, the heart-sound texture feature extraction and identification are realized by the improved pulse-coupled neural network model (IPCNN). Simulation experiments show that compared with the traditional window function, the heart-sound time-frequency diagram obtained using heart-sound window function has a clearer and noise well suppressed texture. Furthermore, compared with other three kinds of typical recognition methods, IPCNN has the lower computational cost and higher recognition rate. So, we can arrive at the conclusion that the method for heart-sound feature extraction and recognition based on image processing techniques is the effective one.
Keywords:heart sounds time-frequency diagram  heart-sounds texture map  heart-sounds window function  improved pulse coupled neural network
点击此处可从《物理学报》浏览原始摘要信息
点击此处可从《物理学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号