Drop fingerprint recognition based on feature extraction and probabilistic neural networks

  • Authors:

  • Qing Song;Jie Li

  • Affiliations:

  • Automation School, Beijing University of Posts and Telecommunications, Beijing, China;Automation School, Beijing University of Posts and Telecommunications, Beijing, China

  • Venue:
  • AICI'11 Proceedings of the Third international conference on Artificial intelligence and computational intelligence - Volume Part III
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Liquid test methods included the authenticity of identification and pattern recognition. Because of the limitation, the authenticity of identification is less useful than pattern recognition. In drop fingerprint pattern recognition, the drop fingerprint data and the eigenvalues of fingerprint are both objects for study. The drop fingerprint feature extraction by waveform analysis can grasp the main features of the liquid, and reduce the amount of data to be processed, greatly improve the recognition efficiency. With the advantage in classification, the Probabilistic Neural Networks provide a good method to the drop fingerprint classification and recognition. By MATLAB simulation, a Probabilistic Neural Networks is established, 8 eigenvalues are set as the pattern stand for a liquid. In the experiment, the rate of correct recognition reach 97.5%.