Study on several methods of character analysis and classification for pulse-condition based on statistics theory .
- Author:
Yue ZHOU
1
;
Qing XU
;
Wei KONG
Author Information
1. Institute of Image Processing & Pattern Recognition, Shanghai Jiaotong University, Shanghai 200030, China. zhouyue@sjtu.edu.cn
- Publication Type:Journal Article
- MeSH:
Humans;
Medicine, Chinese Traditional;
Pulse;
Signal Processing, Computer-Assisted
- From:
Journal of Biomedical Engineering
2006;23(3):505-508
- CountryChina
- Language:Chinese
-
Abstract:
In this paper, several methods of character analysis and classification for pulse condition are studied with the use of statistics theory. Five methods are applied to character abstraction. Pulse condition is analyzed in time, frequency and wavelet domain. The abstraction methods include wavelet analysis, AR modeling, and template match. On the other hand, four classification methods are used, namely BP, RBF, CGA and FMMNN.
