Liquid Recognition Using Fuzzy Pattern Recognition Based on Combined Features

In this paper, droplet fingerprint is taken as the research object. Maximum membership principle classifier which based on fuzzy comprehensive evaluation is used to achieve automatic recognition and classification of liquid. The extraction of eigenvalues, establishment of membership function, weight distribution and the fuzzy decision-making which are the key technologies in fuzzy pattern recognition are carefully discussed in the paper. The experiments proved that the fuzzy pattern recognition system provide a useful reference to improve the liquid identification based on drop fingerprint.