Journal
AD HOC NETWORKS
Volume 113, Issue -, Pages -Publisher
ELSEVIER
DOI: 10.1016/j.adhoc.2020.102360
Keywords
Internet of Things application; Machine learning; Volatile organic compound classification
Funding
- National Research Foundation of Korea (NRF), South Korea - Ministry of Science and ICT (MSIT), South Korea, through the Information Technology Research Center (ITRC) Support Program [IITP-2020-2020-0-01461]
- NRF, South Korea - MSIT [2015R1A5A1037668]
Ask authors/readers for more resources
Improvements in small sized sensors now allow easy detection of volatile organic compounds (VOCs) in the air using IoT devices. However, classifying the types of VOCs in the environment remains complex. VOCkit combines chemically-designed fluorometric sensors with an imaging system for non-chemical experts to detect and classify VOCs easily.
Improvements in small sized sensors allow the easy detection of the presence of Volatile Organic Compounds (VOCs) in the air using easy-to-deploy Internet of Things (IoT) devices. However, classifying what VOC exists in the environment still remains as a complex task. Knowing what VOCs are in the air can help us remove the main cause that vents VOC materials as a way to maintain clean air quality. In this work, we present VOCkit, an IoT sensor kit for non-chemical experts to easily detect and classify different types of VOCs. VOCkit combines miniature chemically-designed fluorometric sensors for recognizing VOCs with an embedded imaging system for classification. Exposing the fluorometric sensors with various VOCs, result in the photophysical property change of fluorescent compounds, which composes the sensors, and the synergistic combination of the changes create unique individual fluorescent color patterns respectively to the VOC material. The fluorescent color change pattern is captured using an embedded camera and the images are processed with machine learning algorithms on the embedded platform for VOC classification. Using 500 fluorometric sensor images collected for five different commonly contactable VOCs, we show the feasibility of VOC classification on small-sized IoT devices. For the VOC types of our interest, our results show a classification accuracy of 97%, implying the potential applicability of VOCkit for real-world usage.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available