Rapid determination of chemical composition and classification of bamboo fractions using visible–near infrared spectroscopy coupled with multivariate data analysis

Table 5 Results of calibration and prediction models for classifying mature bamboo fractions

Wavelength (nm)	Sample sets	Parameters	B	I	K	M	O
400–780 (n = 100)	Calibration	R ²	0.96	0.88	0.81	0.93	0.93
	Calibration	SEC	0.08	0.14	0.17	0.11	0.11
	Validation	R ²	0.90	0.77	0.69	0.71	0.83
	Validation	SEV	0.13	0.19	0.23	0.21	0.17
	Factors		13	13	13	13	13
780–2500 (n = 100)	Calibration	R ²	0.98	0.94	0.93	0.97	0.97
	Calibration	SEC	0.05	0.10	0.10	0.08	0.07
	Validation	R ²	0.93	0.76	0.76	0.85	0.90
	Validation	SEV	0.11	0.20	0.20	0.16	0.13
	Factors		16	16	16	16	16
400–2500 (n = 100)	Calibration	R ²	0.98	0.98	0.96	0.97	0.96
	Calibration	SEC	0.06	0.06	0.08	0.07	0.08
	Validation	R ²	0.94	0.95	0.87	0.90	0.91
	Validation	SEV	0.10	0.09	0.14	0.13	0.12
	Factors		15	15	15	15	15

Three regions of wavelength were selected to establish models, and these spectra came from original data. The number of sample sets was 100. (B bamboo branch, I bamboo yellow, K bamboo node, M bamboo timber, O bamboo green)

ISSN: 2731-3654