DocumentCode
2268640
Title
Pattern Recognition of the Universal Electronic Nose
Author
Tao, Zhou ; Lei, Wang ; Teng, Jionghua
Author_Institution
Coll. of Electron. & Inf. Eng., TongJi Univ., Shanghai
Volume
3
fYear
2008
fDate
20-22 Dec. 2008
Firstpage
249
Lastpage
253
Abstract
An electronic nose is the intelligent instrument that identifies the chemical odors mimicking a human. Now the majority of electronic noses could only identify the specific species, however the human olfactory system is able to characterize and classify many different odors. The problem has prevented their use in wider commercial applications. The pattern recognition methods based on the probabilistic neural networks (PNN) are studied in this paper. The electronic nose systems designed could identify all the samples of beer, fruit juice and milk successfully in the experiments. The results of the experiments showed that the researched systems have a better classification and generalization capacity. The pattern recognition methods of the universal electronic nose are proposed in the paper. The effective universal electronic nose has much advantage over others such as simple methods of pattern recognition and classification, easy training approaches and wider application fields.
Keywords
beverage industry; electronic noses; neural nets; pattern recognition; probability; production engineering computing; beer sample; chemical odors; fruit juice sample; human olfactory system; milk sample; pattern recognition; probabilistic neural networks; universal electronic nose; Chemical analysis; Dairy products; Electronic noses; Food industry; Humans; Instruments; Neural networks; Olfactory; Pattern recognition; Sensor arrays; Electronic nose; Pattern recognition; Probabilistic neural networks; samples;
fLanguage
English
Publisher
ieee
Conference_Titel
Intelligent Information Technology Application, 2008. IITA '08. Second International Symposium on
Conference_Location
Shanghai
Print_ISBN
978-0-7695-3497-8
Type
conf
DOI
10.1109/IITA.2008.416
Filename
4739996
Link To Document