におい感覚量評価を行うセンシングシステム(<小特集>感性のシミュレーション)

概要

論文の詳細を見る
Odor sensing system is required in many fields such as food, drink, cosmetics and environmental testing etc. since there has been no objective evaluation method instead of traditional human sensory test. In the present study, sensing system, by which human sensory test can be replaced, was studied for predicting odor sensory quantities. It is composed of vapor supply system using auto-sampling stage, sensors, electronic portion and neural network. Neural network can be used for mapping sensor data space onto sensory quanity space after training. The experiment was performed so that odor similatities could be extracted from the data of a quartz-resonator sensor array by FLVQ (Fuzzy Learning Vector Quantization, a neural network model proposed by the authors). It was found that the odor similarities among various kinds of flavors such as bergamot and petitgrain etc. could be estimated using that system.
日本シミュレーション学会の論文
1994-03-15