Journal of Analytical Science and Technology

Table 3 Predictive results of samples using RGB color model after normalization

From: Investigation into the predictive performance of colorimetric sensor strips using RGB, CMYK, HSV, and CIELAB coupled with various data preprocessing methods: a case study on an analysis of water quality parameters

	Mn²⁺ (ppm)			Fe²⁺/Fe³⁺ (ppm)
Sample	[Spiked sample]	RGB + normalization	%Recovery	[Spiked sample]	RGB + normalization	%Recovery
W1	40	39.85 ± 1.91	95–104	40	57.82 ± 3.39	136–153
W2	40	31.74 ± 1.28	76–82	40	65.09 ± 2.78	156–170
W3	40	35.31 ± 7.05	71–106	40	61.30 ± 4.22	143–164
W4	40	40.45 ± 0.56	100–103	40	67.24 ± 3.19	160–176
	Cu²⁺ (ppm)			NO₃^- (ppm)
Sample	[Spiked sample]	RGB + normalization	%Recovery	[Spiked sample]	RGB + normalization	%Recovery
W1	150	160.27 ± 6.87	102–111	250	292.47 ± 33.37	104–130
W2	150	147.98 ± 4.89	95–102	250	297.02 ± 24.57	109–129
W3	150	150.93 ± 3.42	98–103	250	286.75 ± 16.77	108–121
W4	150	152.57 ± 10.15	95–108	250	260.39 ± 23.95	95–114
	PO₄^3- (ppm)			SO₄^2- (ppm)
Sample	[Spiked sample]	RGB + normalization	%Recovery	[Spiked sample]	RGB + normalization	%Recovery
W1	40	41.51 ± 3.06	77–89	800	813.24 ± 180.29	79–124
W2	40	59.79 ± 1.39	106–112	800	860.69 ± 108.44	94–121
W3	40	35.14 ± 1.83	67–74	800	848.61 ± 60.37	98–113
W4	40	36.94 ± 3.07	68–80	800	932.06 ± 13.90	115–118
	Total hardness (CaCO₃, ppm)			pH
Sample	[Sample]	RGB + normalization	%Error	pH	RGB + normalization	%Error
W1	93.3	62.82 ± 19.96	11–54	7.43	7.06 ± 0.11	3–6
W2	189.3	183.95 ± 18.25	7–12	7.96	6.11 ± 0.13	22–25
W3	85.3	61.03 ± 18.25	7–5	7.32	7.84 ± 0.07	8–6
W4	24	16.90 ± 26.97	83–142	7.32	6.72 ± 0.33	4–13

Back to article page