The utility of a near-infrared spectrometer to predict the maturity of green peas (Pisum sativum)

•The standard error of the laboratory (SEL) for a tenderometer was observed to be 1.1 tenderometer (TR) units.•Tenderometer reading had no relationship (R = −0.16) with pea size and a weak negative linear relationship (R = −0.44) with pea moisture content.•A single-season calibration was developed with SVM regression over a 3-month harvest period of successively planted peas without controlling the physical form or temperature resulting in an RMSEC/CV/P = 3.8/4.7/4.4 TR, R2C/CV/P = 0.86/0.78/0.83 and an RPD of 2.4 over a range of TR from 65 to 104.•Inspection of the loadings for the first four latent variables indicates that a spectral range beyond 1050 nm has utility in the prediction of tenderometer reading. This work evaluated the utility of an embedded diode array near-infrared reflectance spectrometer to predict the hardness (maturity) of green peas. Partial least squares, support vector machine (SVM) and artificial neural network regression models were developed to predict tenderometer reading (TR) from near-infrared spectra over the range of maturities experienced throughout a harvest season. The SVM model achieved the best performance in an independent data set collected in a second harvest season. Here, the SVM model explained 83% of the variation in TR with a root mean standard error of prediction (RMSEP) of 4.4 TR units over a range of TR from 65 to 104. The RPD of 2.3 achieved in this study exceeds the threshold for rough screening but falls short of replacing the tenderometer for quality control.