Water Quality Assessment with Thermal Images
Loading...
Links to Files
Author/Creator
Author/Creator ORCID
Date
2020-09
Type of Work
Department
Program
Citation of Original Publication
N. Khan and N. Roy, "Water Quality Assessment with Thermal Images," 2020 IEEE International Conference on Smart Computing (SMARTCOMP), Bologna, Italy, 2020, pp. 164-171, doi: 10.1109/SMARTCOMP50058.2020.00041.
Rights
This item is likely protected under Title 17 of the U.S. Copyright Law. Unless on a Creative Commons license, for uses protected by Copyright Law, contact the copyright holder or the author.
© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Subjects
Abstract
Water contamination has been a critical issue in many countries of the world including USA. Physical, chemical, biological, radio-logical substances can be the reason of this contamination. Drinking water systems are allowed to contain chlorine, calcium, lead, arsenic etc., at a certain level. However, there are expensive instruments and paper sensors to detect the quantity of minerals in water. But these instruments are not always convenient for easy determination of the quality of the sample as drinking water. Different minerals in the water reacts to heat heterogeneously. Some minerals (i.e., arsenic) stay in the water with noticeable amount even after reaching to boiling point. However, it requires cheaper and easier process to examine the quality of water samples for drinking from different sources. With this in mind, we experimented few water samples from different places of USA including artificially prepared samples by mixing different impurities. We investigated their heating property with the sample of marked safe drinking water. We collected thermal images with 10-seconds interval during cooling period of hot water samples from the boiling point to room temperature. We extracted features for each of the water samples with the combination of convolution and recurrent neural network based model and classified different water samples based on the added impurity types and sources from where the samples were collected. We also showed the feature distances of these water samples with the safe water sample. Our proposed framework can differentiate features for different impurities added in the water samples and detect different category of impurities with average accuracy of 70%.