Table 2.

Overall and individual class classification accuracies of machine-learning models with texture features. ^a

Classifier	Parameter	Calibration set accuracy (%)					Prediction set accuracy (%)
		Healthy	2 DPI	4 DPI	6 DPI	Overall	Healthy	2 DPI	4 DPI	6 DPI	Overall
PLS-DA	2	85.00	95.00	50.00	60.00	76.67	90.00	100.00	60.00	50.00	80.00
RF	75	100.00	100.00	100.00	100.00	100.00	93.33	50.00	80.00	100.00	85.00
SVM	(0.01, 3.03)	100.00	100.00	100.00	100.00	100.00	96.67	60.00	70.00	100.00	86.67
LS-SVM	(1.48, 22.34)	100.00	90.00	90.00	100.00	96.67	100.00	30.00	70.00	100.00	83.33
ELM	42	100.00	95.00	90.00	100.00	97.50	93.33	90.00	70.00	100.00	90.00
BPNN	10	98.33	85.00	90.00	100.00	95.00	96.67	90.00	80.00	100.00	93.33

^aParameters and abbreviations as in Table 1.