We present an Electronic Nose aimed at identifying a gas. Our integrated hardware-software system uses some machine learning principles to identify a new gas sample. It is based on eight sensors of different types and their output data are processed by using a Support Vector Machine (SVM) classifier. The system has been tested on four different types of gases (Methanol, Ethanol, Acetone and Benzene) for different concentrations. Both binary classification for each pair of gases and multiclassification for the whole set of gases have been performed. Classification of mixtures has been performed as well. The results of the experiments are presented, together with the architecture of the system.
SVM classification in the design of an electronic nose
Gaudioso M;PACE, Calogero
2007-01-01
Abstract
We present an Electronic Nose aimed at identifying a gas. Our integrated hardware-software system uses some machine learning principles to identify a new gas sample. It is based on eight sensors of different types and their output data are processed by using a Support Vector Machine (SVM) classifier. The system has been tested on four different types of gases (Methanol, Ethanol, Acetone and Benzene) for different concentrations. Both binary classification for each pair of gases and multiclassification for the whole set of gases have been performed. Classification of mixtures has been performed as well. The results of the experiments are presented, together with the architecture of the system.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.