Skip to main content

Table 2 Statistics in calibration and prediction sets for M IAM or M CS property from NIR or SWIR HSI spectra

From: Feasibility study of assessing cotton fiber maturity from near infrared hyperspectral imaging technique

Maturity & optimal factora

Calibration set

Prediction set

Range

SDb

R C 2

RMSECb

No./ %c

Range

SDb

R P 2

RMSEPb

No. / %c

M IAM, 5

0.570∼1.087

0.099

0.828

0.041

68 / 97.1

0.632∼1.030

0.085

0.739

0.044

33 / 97.1

M CS, 8

0.530∼0.970

0.076

0.835

0.036

66 / 98.5

0.650∼0.947

0.064

0.518

0.046

32 / 97.0

  1. aOptimal models from spectral preprocessing with combinations of mean centering (MC), multiplicative scatter correction (MSC), and Savitzky-Golay (SG) first derivative (FD) function of thirteen points; For M IAM, 70 samples in calibration set and 34 samples in prediction set; For M CS, 67 samples in calibration set and 33 samples in prediction set.
  2. bStandard deviation (SD), root mean square error of calibration (RMSEC) and prediction (RMSEP).
  3. cNumber and percentage (%) of samples with the difference between predicted and referenced value within the 95% agreement range of bias ± 1.96*RMSEC (or RMSEP) in calibration and prediction set.
Back to article page

Journal of Cotton Research

ISSN: 2523-3254

Contact us