The best performing color space and machine learning regression algorithm for the accurate estimation of chromium (VI) and iron (III) in aqueous samples using low-cost and portable flatbed scanner colorimetry

The study utilizes the colorimetric method (involving 1,5-diphenylcarbazide and potassium thiocyanate as complexing agents), computer vision, and machine learning (ML) regression algorithms to determine the content of Cr (VI) and Fe (III) in water samples. To process digital images of water samples, the integration technique utilized a flatbed scanner known as the CanoScan LiDE 100, operating as a digital image capture device, and its performance was compared to that of conventional instruments. The study reveals that PolyReg and SVR-Poly are the most reliable ML regression algorithms for processing color space data (G and B of RGB, c* of CIELch, and b* of CIELab) of digital images of water samples that contain Cr (VI) and Fe (III). The mean absolute percentage error (MAPE) of the ML regression algorithms PolyReg and SVR-Poly for determining the content of Cr (VI) and Fe (III) is