Abstract
Gas identification represents a big challenge for pattern recognition systems due to several particular problems. The aim of this study is to compare the accuracy of a range of advanced and classical pattern recognition algorithms for gas identification from sensor array signals. Density estimation is applied in the construction of classifiers through the use of Bayes rule. Experiments on real sensors' data proved the effectiveness of the approach with an excellent classification performance. We compare the classification accuracy of different density models with several neural networks architectures. On our gas sensors data, the best performance was achieved by Gaussian mixture models with more than 92% accuracy.
Original language | English |
---|---|
Pages (from-to) | 584-587 |
Number of pages | 4 |
Journal | Conference Record - IEEE Instrumentation and Measurement Technology Conference |
Volume | 1 |
Publication status | Published - 8 Oct 2004 |
Externally published | Yes |
Event | Proceedings of the 21st IEEE Instrumentation and Measurement Technology Conference, IMTC/04 - Como, Italy Duration: 18 May 2004 → 20 May 2004 |
Keywords
- Classification
- Density models
- Gas sensor array
- Neural networks
- Pattern recognition
ASJC Scopus subject areas
- Instrumentation