Skip to main content

Table 1 Results of PLS1 calibration and prediction models for the quantitative compositional analysis of bamboo using raw spectra

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

Wavelength (nm)

Chemical composition

Factors

R2 c

RMSEC

R2 p

RMSEP

SD

RPD

RER

400–780 (n = 36)

Cellulose

8

0.82

4.3

0.68

6.0

10.3

1.7

5.6

Xylan

3

0.87

3.0

0.83

3.5

8.4

2.4

8.0

Lignin

3

0.83

3.7

0.78

4.2

8.9

2.1

7.1

Glucose

3

0.79

8.2

0.70

9.7

18.2

1.9

5.4

Xylose

10

0.97

4.3

0.86

9.0

23.7

2.6

8.7

780–2500 (n = 36)

Cellulose

4

0.88

3.5

0.82

4.3

10.3

2.4

6.5

Xylan

6

0.93

2.1

0.88

2.9

8.4

2.9

10.0

Lignin

7

0.96

1.8

0.92

2.5

8.9

3.6

11.9

Glucose

7

0.92

5.1

0.77

8.6

18.2

2.1

7.6

Xylose

4

0.90

7.5

0.79

11.1

23.7

2.1

6.0

400–2500 (n = 36)

Cellulose

4

0.88

3.5

0.82

4.3

10.3

2.4

6.1

Xylan

4

0.94

2.1

0.91

2.5

8.4

3.3

10.0

Lignin

7

0.94

2.2

0.86

3.3

8.9

2.7

8.7

Glucose

4

0.84

7.1

0.73

9.4

18.2

1.9

5.9

Xylose

5

0.87

8.4

0.75

11.7

23.7

2.0

7.0

  1. R 2 c square of the correlation coefficient for calibration, RMSEC root mean square error of calibration, R 2 p square of the correlation coefficient for prediction, RMSEP root mean square error of prediction, SD standard deviation, RPD ratio of root mean square error of prediction to standard deviation, RER range error ratio. The number of samples used for quantitative analysis of cellulose, xylan, and lignin is 36. The number of samples used for quantitative analysis of glucose and xylose is 26