Multivariate calibration in smartphones using digital images for the determination of pH of drinking water

The execution of calibration models was previously restricted to computers, but with technological advances, applications on mobile devices have made this process more accessible. Photometrix PRO stands out as an innovative tool for creating calibration models from digital images. The colorimetric chart, widely used to determine the pH of solutions by visual analysis, can present subjectivity in interpretation. This study aims to develop and evaluate calibration models (PLS) using Photometrix PRO to accurately correlate the visual information of a colorimetric pH chart from the company AlfaKit. The application was used to acquire images, generate calibration curves and obtain pH values from prediction and validation sets. The curves presented R2, RMSEC and RMSECV that indicated good fit of the models. The t-test, with 95% confidence, showed that there was no significant difference between the results of the models and the colorimetric method. The Residual Prediction Deviation (RPD) ranged from 1.12 (RGB) to 5.37 (RGBHSVLI), making the model (RGBHSVLI) suitable for quality control. The study successfully demonstrated the feasibility and effectiveness of smartphone-based multivariate calibration (PLS) for determining pH in drinking water.