Computer Vision Technique to Improve the Color Ratio in Estimating the Concentration of Free Chlorine

Ruiz-Navarro J.A.; Santos-López F.M.; Portella-Delgado J.M.; Santos-de-la-Cruz E.G. · Lecture Notes in Networks and Systems (2022) · DOI: 10.1007/978-3-030-97719-1_7

Resumen

Access to drinkable water is a constitutional right in Peru. However, quality water is not universally accessible, particularly in rural areas where neither water disinfection nor quality control are undertaken. It is therefore crucial to develop methods that allow estimates of water quality characteristics, such as free chlorine concentration. This publication proposes a sensing system that estimates free chlorine concentration using a traditional colorimetric method of chemical reaction with N, N-diethyl-p-phenylenediamine and a stationary digital camera. Using image processing, a subset of computer vision, the algorithm quantifies the color of the water sample after chemical reaction with the reagent and, on the basis of that quantification, tries to estimate the free chlorine concentration of the sample. After the quantification, a model is determined using a best-fit polynomial. The resulting polynomial model is then implemented using serverless cloud computing resources from Amazon Web Services, and its results are also hosted using this platform. The model error has a standard deviation of 0.0129 ppm.

Instituciones: Universidad Nacional Mayor de San Marcos; Pontificia Universidad Católica del Perú