Multichannel Odor Sensor System using Chemosensitive Resistors and Machine Learning

Atsushi Shunori, Rui Yatabe, Bartosz Wyszynski, Yosuke Hanai, Atsuo Nakao, Masaya Nakatani, Akio Oki, Hiroaki Oka, Takashi Washio, Kiyoshi Toko

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

In this study, we have fabricated multichannel odor sensor using chemosensitive resistors. The chemosensitive resistors were made from complex of carbon black and gas chromatography stationary materials (GC materials). The electrical resistance of the chemosensitive resistor changed by sensing gas species. We have fabricated an odor sensor chip, which had 16 types of chemosensitive resistors. In addition, we developed a measurement instrument with compact size. The odor sensor chip was embedded in the instrument to construct an odor sensor system. The sensor system outputted the data of 16 channels if sensing gas species. The data have been analyzed using machine learning algorithms that were available on software Weka. As a result, it was successful to identify alcohol beverages by sensing their odor using the sensor system.

Original languageEnglish
Title of host publicationISOEN 2019 - 18th International Symposium on Olfaction and Electronic Nose, Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538683279
DOIs
Publication statusPublished - May 2019
Event18th International Symposium on Olfaction and Electronic Nose, ISOEN 2019 - Fukuoka, Japan
Duration: May 26 2019May 29 2019

Publication series

NameISOEN 2019 - 18th International Symposium on Olfaction and Electronic Nose, Proceedings

Conference

Conference18th International Symposium on Olfaction and Electronic Nose, ISOEN 2019
Country/TerritoryJapan
CityFukuoka
Period5/26/195/29/19

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Instrumentation

Fingerprint

Dive into the research topics of 'Multichannel Odor Sensor System using Chemosensitive Resistors and Machine Learning'. Together they form a unique fingerprint.

Cite this