Encoded Image-Based Time Series Classification for Improving Colorimetric Detection of Hydrogen Sulfide (H2S)

Chang Hyun Kim, Junyeop Lee, Junkyu Park, Daewoong Jung, Chang Woo Nam, Yuntae Ha, Kwan Woo Kim, Sang Hyeok Park, Su Ji Choi, Sanghun Choi, Suwoong Lee

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this study, a time series data analysis technique using a convolutional neural network (CNN), that performs multidimensional image encoding, is used to improve the accuracy of hydrogen sulfide gas detection. According to a recent study, the time -series data image-encoding technique is effective under specific conditions. The novelty of this study lies in the use of a time-series-based colorimetric analysis method developed using colorimetric fabric detection data from sensors to estimate hydrogen sulfide gas exposure levels. Time series data obtained through gas experiments are image-encoded to classify the color value change trend of a dyed fabric induced by its chemical reaction with hydrogen sulfide gas. The results show that learning using encoded image training data improves model performance in estimating gas exposure levels compared to the non-encoded image method.

Original languageEnglish
Title of host publication2022 IEEE Sensors, SENSORS 2022 - Conference Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665484640
DOIs
StatePublished - 2022
Event2022 IEEE Sensors Conference, SENSORS 2022 - Dallas, United States
Duration: 30 Oct 20222 Nov 2022

Publication series

NameProceedings of IEEE Sensors
Volume2022-October
ISSN (Print)1930-0395
ISSN (Electronic)2168-9229

Conference

Conference2022 IEEE Sensors Conference, SENSORS 2022
Country/TerritoryUnited States
CityDallas
Period30/10/222/11/22

Keywords

  • colorimetric analysis
  • gas sensor
  • hydrogen sulfide detection
  • image encoding
  • time-series

Fingerprint

Dive into the research topics of 'Encoded Image-Based Time Series Classification for Improving Colorimetric Detection of Hydrogen Sulfide (H2S)'. Together they form a unique fingerprint.

Cite this