Table 2. Prediction model results of black soldier fly larvae proximate contents using the SWIR hyperspectral imaging system

Parameter	Preprocessing	Whole insect sample
		Rc2	RMSEC (%)	Rp2	RMSEP (%)	LV
Moisture	Mean norm	0.97	1.80	0.98	1.92	5
	Max norm	0.97	1.79	0.98	1.83	5
	Range norm	0.95	2.22	0.96	2.44	4
	MSC	0.96	2.00	0.97	2.07	4
	SNV	0.96	1.97	0.97	2.05	5
	SG 1st	0.94	2.55	0.96	2.41	5
	SG 2nd	0.93	2.71	0.96	2.59	3
	Raw	0.94	2.59	0.96	2.46	5
Crude protein	Mean norm	0.98	0.59	0.98	0.57	4
	Max norm	0.98	0.58	0.99	0.55	5
	Range norm	0.97	0.73	0.97	0.78	4
	MSC	0.98	0.62	0.98	0.61	4
	SNV	0.98	0.61	0.98	0.59	5
	SG 1st	0.96	0.92	0.95	0.98	5
	SG 2nd	0.96	0.93	0.95	0.99	3
	Raw	0.96	0.92	0.95	0.97	5
Crude fat	Mean norm	0.91	1.34	0.91	1.34	4
	Max norm	0.90	1.44	0.90	1.41	4
	Range norm	0.89	1.47	0.90	1.44	4
	MSC	0.91	1.36	0.91	1.39	4
	SNV	0.91	1.37	0.91	1.38	5
	SG 1st	0.88	1.57	0.88	1.61	4
	SG 2nd	0.87	1.60	0.88	1.61	3
	Raw	0.87	1.63	0.87	1.67	4
Crude fiber	Mean norm	0.87	0.45	0.89	0.46	16
	Max norm	0.87	0.46	0.89	0.46	17
	Range norm	0.87	0.46	0.89	0.46	17
	MSC	0.85	0.48	0.85	0.53	15
	SNV	0.85	0.49	0.85	0.52	16
	SG 1st	0.87	0.46	0.86	0.51	14
	SG 2nd	0.86	0.47	0.85	0.53	14
	Raw	0.86	0.48	0.86	0.51	17
Crude ash	Mean norm	0.95	0.24	0.96	0.25	4
	Max norm	0.96	0.23	0.96	0.25	5
	Range norm	0.94	0.27	0.94	0.30	4
	MSC	0.95	0.25	0.96	0.26	4
	SNV	0.95	0.24	0.95	0.27	5
	SG 1st	0.93	0.29	0.95	0.30	5
	SG 2nd	0.92	0.30	0.94	0.32	3
	Raw	0.92	0.30	0.94	0.30	5

SWIR, short wavelength infrared hyperspectral imaging system; Mean norm, mean normalization; Maximum norm, maximum normalization; Range norm, range normalization; MSC, multiplicative scatter correction; SNV, regular normal variate; SG 1^st, Savitzky-Golay 1^st derivation; SG 2^nd, Savitzky-Golay 2^nd derivation; Raw, raw spectrum; R_C², coefficient of determination of calibration set; RMSEC, root mean square error of calibration set; R_P², coefficient of determination of prediction set; RMSEP, root mean square error of prediction set; LV, latent variables.