Using Field Asymmetric Ion Mobility Spectrometry for Odor Assessment of Automobile Interior Components

The of the quality of odors emitted from automobile cabin interiors is an important element for the design of vehicles that meet prospective customers' expectations. Extending our previous work on machine-versus-human odor assessment for intact automobile cabin interiors, in this paper, we eval...

Full description

Saved in:
Bibliographic Details
Published in:IEEE sensors journal 2016-07, Vol.16 (14), p.5747-5756
Main Authors: Juan Li, Gutierrez-Osuna, Ricardo, Hodges, Ryan D., Luckey, Gail, Crowell, Joel, Schiffman, Susan S., Nagle, H. Troy
Format: Article
Language:English
Subjects:
Assessments
Automobile industry
Automobile odor assessment
Automobiles
Automotive components
Automotive engineering
Chemicals
Compounds
Feature extraction
field asymmetric ion mobility spectrometry
image processing
Ionic mobility
Ionization
Ions
k-fold cross-validation
machine olfaction
Mathematical models
Odors
Panels
partial least squares regression
Sensors
Spectroscopy
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The of the quality of odors emitted from automobile cabin interiors is an important element for the design of vehicles that meet prospective customers' expectations. Extending our previous work on machine-versus-human odor assessment for intact automobile cabin interiors, in this paper, we evaluated odors generated from individual interior parts using a human panel and field asymmetric ion mobility spectrometry (FAIMS). We used image processing techniques to extract geometric features from FAIMS dispersion fields, and built the predictive models for three odor assessment parameters (intensity, irritation, and pleasantness) by means of partial least squares regression. The best feature set was chosen by backward sequential feature selection. Using k -fold cross validation, we achieved statistically significant correlation 0.95 between human panel measured and machine olfaction predicted odor assessment scores with a sample set of 48 interior automobile parts. These results, generated using the geometric image processing methods demonstrated herein, further support the feasibility of replacing a human panel by machine olfaction for the assessment of odor quality of interior automobile parts.
ISSN:1530-437X
1558-1748
DOI:10.1109/JSEN.2016.2568209