2025-05-30T17:52:49Z
2025-05-30T17:52:49Z
2025-05-22
2025-05-30T17:52:49Z
Free chlorine (FC) plays a crucial role in ensuring the safety of drinking water by effectively inactivating pathogenic microorganisms. However, traditional methods for measuring FC levels often require specialized equipment and laboratory settings, limiting their accessibility and practicality for on-site or point-of-use monitoring. QR Codes are powerful machine-readable patterns that are used worldwide to encode information (i.e., URLs or IDs), but their computer vision features allow QR Codes to act as carriers of other features for several applications. Often, this capability is used for aesthetics, e.g., embedding a logo in the QR Code. In this work, we propose using our technique to build back-compatible Color QR Codes, which can embed dozens of colorimetric references, to assist in the color correction to readout sensors. Specifically, we target two well-known products in the HORECA (hotel/restaurant/café) sector that qualitatively measure chlorine levels in samples of water. The two targeted methods were a BTB strip and a DPD powder. First, the BTB strip was a pH-based indicator distributed by Sensafe®, which uses the well-known bromothymol blue as a base-reactive indicator; second, the DPD powder was a colorimetric test distributed by Hach®, which employs diethyl-p-phenylenediamine (DPD) to produce a pink coloration in the presence of free chlorine. Custom Color QR Codes were created for both color palettes and exposed to several illumination conditions, captured with three different mobile devices and tested over different water samples. Results indicate that both methods could be correctly digitized in real-world conditions with our technology, rendering a 88.10% accuracy for the BTB strip measurement, and 84.62% for the DPD powder one.
Artículo
Versión publicada
Inglés
Aprenentatge automàtic; Colorimetria; Clor; Machine learning; Colorimetry; Chlorine
MDPI
Reproducció del document publicat a: https://doi.org/10.3390/s25113251
Sensors, 2025
https://doi.org/10.3390/s25113251
cc-by (c) González-Gómez, M. et al., 2025
http://creativecommons.org/licenses/by/4.0/