Title :
Odor Recorder Using Mass Spectrometry and Large-scale Data
Author :
Nakamoto, Takamichi ; Nakama, Tomomasa
Author_Institution :
Tokyo Inst. of Technol. 2, Tokyo
Abstract :
Since a variety of smells can be generated by blending multiple odor components, an odor recorder can reproduce smells as well as record them. Although the range of smell to be recorded has been so far limited this work enhances that range using mass spectrometry without GC. We proposed the algorithm to select appropriate odor components among huge number of candidates. In the simulation, the recipe of smell was correctly obtained using odor components selected among 10,000 candidates. In the experiment, the apple flavor was successfully blended using odor components selected among 82 candidates. Furthermore, sensory test revealed that the smell blended according to the estimated recipe was almost the same as the target one.
Keywords :
electronic noses; mass spectroscopy; apple flavor; large-scale data; mass spectrometry; odor components; odor recorder; recipe smell; sensory test; Constraint theory; Data engineering; Databases; Educational institutions; Gas detectors; Large-scale systems; Least squares methods; Mass spectroscopy; Sensor arrays; Testing;
Conference_Titel :
Sensors, 2007 IEEE
Conference_Location :
Atlanta, GA
Print_ISBN :
978-1-4244-1261-7
Electronic_ISBN :
1930-0395
DOI :
10.1109/ICSENS.2007.4388364