Unknown odor recognition using Euclidean Fuzzy similarity-based Self-Organized Network inspired by Immune Algorithm |
| |
Authors: | Muhammad R Widyanto Benyamin Kusumoputro Kaoru Hirota |
| |
Affiliation: | (1) Faculty of Computer Science, University of Indonesia, Depok Campus, Depok, 16424, West Java, Indonesia;(2) Department of Computational Intelligence and Systems Science, Tokyo Institute of Technology, G3-49, 4259 Nagatsuta, Midori, Yokohama 226-8502, Japan |
| |
Abstract: | To deal with unknown odor recognition problem for a developed artificial odor discrimination system, Euclidean Fuzzy similarity-based
Self-Organized Network inspired by Immune Algorithm (EF-SONIA) is proposed. Euclidean fuzzy similarity enables a zero similarity
calculation between an unknown odor vector and hidden unit vectors, so that the system can recognize the unknown odor. In
addition, an elliptical approach for fuzziness determination is proposed. The elliptical approach can approximate an appropriate
fuzziness, so that the unknown odor recognition accuracy is improved. Experiments on three datasets of three-mixture vegetal
odors show that the recognition accuracy of the proposed method is 20% better than those of the conventional method. The system
is very promising to be used for a real development of dog robot that enables localization and identification of dangerous
natural gas. |
| |
Keywords: | Odor discrimination Self-organization Immune algorithm Fuzzy similarity Euclidean distance |
本文献已被 SpringerLink 等数据库收录! |
|