Title :
A Calculating Method of Similarity Degree for Fuzzy Pattern Recognition
Author :
Zhao, Quanming ; Li, Lingling ; Li, Zhigang ; Gao, Zhaohui ; Wang, Jiannan
Author_Institution :
Dept. of Electron. Inf., Tianjin Univ.
Abstract :
In order to avoid error recognition in fuzzy pattern recognition, a new algorithm of closeness degree was proposed to calculate the similarity on a certain characteristic between the typical pattern and the pattern to be recognized, then a function of similarity degree was established. This algorithm featured in combining the membership degree with closeness degree by distributing different weights and ability to process pattern eigenvalue represented by real number or fuzzy set, adaptive adjusting weights to each closeness degree in terms of pattern eigenvalue so as to implement correct pattern recognition. The validity of this algorithm was approved by two examples
Keywords :
eigenvalues and eigenfunctions; fuzzy set theory; pattern recognition; fuzzy pattern recognition; fuzzy set; pattern eigenvalue; similarity degree calculation; Automation; Character recognition; Computer errors; Eigenvalues and eigenfunctions; Electromagnetic fields; Fuzzy sets; Intelligent control; Pattern recognition; closeness degree; fuzzy pattern recognition; membership degree; similarity degree function;
Conference_Titel :
Intelligent Control and Automation, 2006. WCICA 2006. The Sixth World Congress on
Conference_Location :
Dalian
Print_ISBN :
1-4244-0332-4
DOI :
10.1109/WCICA.2006.1713003
