Yamanaka Takao

The smell reproduction technique in addition to that for vision and hearing will be useful in the fields of virtual environments, games and e-commerce. In this study, citrus flavors were electronically recorded using. "odor recorder" composed of quartz crystal microbalance odor sensors. First, an artificial orange flavor composed of 14 constituents was successfully recorded with its approximated recipe of 3 odor components. The human sensory test revealed, that the reproduced odor was close to the target odor for the human olfaction. Next, we tried to record natural smells from citrus essential oils. The mixture smells of two essential oils were successfully recorded using the odor recorder.

The smell reproduction technique in addition to those for vision and hearing will be useful in the fields of virtual environments, games and e-commerce in fragrances and foods. We have developed an odor recorder to record odors using quartz crystal microbalance sensors and to reproduce them based on the recorded data. In this paper, the results of recording dynamical changes in odors in the atmosphere are described. The dynamical changes in the odors were successfully recorded in spite of the rapid change of the odor concentration over a few seconds due to the airflow turbulence. First, the change in a binary mixture of apple-flavor components was successfully recorded using the odor recorder based on a feedback control. Next, the changes in four fruit flavors, apple, mandarin, pineapple, and banana flavors, were successfully recorded using the odor recorder based on a pattern classification method.

The new method for the odor recorder, electronically recording the recipe of odors or scents made up of many components, was proposed. Although the apple flavors have been recorded with the mixture of five components using the odor recorder in previous work, the number of odor components should be increased to expand the applicable range of odors. Since the collinearity problem of the odor sensor array became apparent with the increase of the components, the new method based on the variable transformation using SVD (Singular Value Decomposition) was developed in this paper to extract the effective subspace of the sensor outputs for the recipe exploration. As a result, the sensor-array response pattern of the reproduced odor with the 8-component recipe almost agreed with that of the target apple flavor. Furthermore, the human sensory tests revealed that the smell of the approximated odor was identical to that of the target flavor.