华南理工大学学报(自然科学版) ›› 2007, Vol. 35 ›› Issue (1): 49-52.

• 电子、通信与自动控制 • 上一篇    下一篇

基于第二代小波变换的中药指纹图谱的基线提取

杨云 朱学峰   

  1. 华南理工大学 自动化科学与工程学院,广东 广州 510640
  • 收稿日期:2005-11-22 出版日期:2007-01-25 发布日期:2007-01-25
  • 通信作者: 杨云(1973-) ,女,博士生,主要从事模式识别、专家系统方面的研究。 E-mail:yangwmy@yahoo.com.cn
  • 作者简介:杨云(1973-) ,女,博士生,主要从事模式识别、专家系统方面的研究。
  • 基金资助:

    广州市科技攻关项目(2004Z3-E5011 )

Baseline Spliting of Traditional Chinese Medicine FingerprintBased on the Second-Generation Wavelet Transform

Yang Yun  Zhu Xue-feng   

  1. School of Automation Science and Engineering , South China Univ. of Tech. , Guangzhou 510640 , Guangdong , China
  • Received:2005-11-22 Online:2007-01-25 Published:2007-01-25
  • Contact: 杨云(1973-) ,女,博士生,主要从事模式识别、专家系统方面的研究。 E-mail:yangwmy@yahoo.com.cn
  • About author:杨云(1973-) ,女,博士生,主要从事模式识别、专家系统方面的研究。
  • Supported by:

    广州市科技攻关项目(2004Z3-E5011 )

摘要: 对传统的中药指纹图谱信号数据预处理技术存在的算法复杂和运算速度慢的不足,在具体分析中药指纹图谱信号数据特点的基础上,提出利用第二代小波变换和局部重构技术,采用Daubechies 小波提升法来实现中药指纹图谱信息分析过程中的基线提取.对田基黄药材指纹图谱(符合AIA 标准)数据的处理结果表明,该方法可以很好地实现指纹图谱信号峰与基线的分离,提取出的基线更接近真实基线,而且易于软件实现。

关键词: 小波变换, 小波提升, 中药指纹图谱, 基线校正

Abstract:

In order to overcome the disadvantages of the conventional data pre-processing technology for the fingerprintsignals of traditional Chinese medicine , such as complexity and low speed of the algorithm , the features of fingerprintsignals are analyzed , on the basis of which a new method is then proposed. In this method , the secondgenerationwavelet transform technology , the local reconstruction technology and the Daubechies wavelet-liftingmethod are adopted to split the baseline of traditional Chinese medicine fingerprint. The experimental results ofTianjihuang fingerprint according with the AIA standard show that the proposed method can effectively split thebaseline from the signal peak with great precision and is easy to be implemented by software.

Key words: wavelet transform, wavelet lifting, traditional Chinese medicine fingerprint, baseline correction