Multimodal Imaging and Lighting Bias Correction for Improved μPAD-based Water Quality Monitoring via Smartphones
AffiliationUniv Arizona, Dept Agr & Biosyst Engn
Univ Arizona, Mel & Enid Zuckerman Coll Publ Hlth
MetadataShow full item record
PublisherNATURE PUBLISHING GROUP
CitationMultimodal Imaging and Lighting Bias Correction for Improved μPAD-based Water Quality Monitoring via Smartphones 2016, 6:27529 Scientific Reports
RightsThis work is licensed under a Creative Commons Attribution 4.0 International License.
Collection InformationThis item from the UA Faculty Publications collection is made available by the University of Arizona with support from the University of Arizona Libraries. If you have questions, please contact us at email@example.com.
AbstractSmartphone image-based sensing of microfluidic paper analytical devices (mu PADs) offers low-cost and mobile evaluation of water quality. However, consistent quantification is a challenge due to variable environmental, paper, and lighting conditions, especially across large multi-target mu PADs. Compensations must be made for variations between images to achieve reproducible results without a separate lighting enclosure. We thus developed a simple method using triple-reference point normalization and a fast-Fourier transform (FFT)-based pre-processing scheme to quantify consistent reflected light intensity signals under variable lighting and channel conditions. This technique was evaluated using various light sources, lighting angles, imaging backgrounds, and imaging heights. Further testing evaluated its handle of absorbance, quenching, and relative scattering intensity measurements from assays detecting four water contaminants-Cr(VI), total chlorine, caffeine, and E. coli K12-at similar wavelengths using the green channel of RGB images. Between assays, this algorithm reduced error from mu PAD surface inconsistencies and cross-image lighting gradients. Although the algorithm could not completely remove the anomalies arising from point shadows within channels or some non-uniform background reflections, it still afforded order-of-magnitude quantification and stable assay specificity under these conditions, offering one route toward improving smartphone quantification of mu PAD assays for in-field water quality monitoring.
VersionFinal published version
SponsorsNational Science Foundation Graduate Research Fellowship [DGE-1143953]; Paul D. Coverdell Fellows Program; Water and Environmental Technology (WET) Center at the University of Arizona; Tucson Water