Near-infrared spectroscopy and accelerated ageing in evaluating the vigour of lentil seeds

Abstract

The cultivation and consumption of lentils has been gaining importance in recent years due to the high
nutritional value of the grain. Research related to seed technology is therefore important in providing high-quality seeds for the
market. The aim o f this study was to adapt the accelerated ageing test and evaluate the potential of FT-NIR spectroscopy for
classifying lentil seeds based on physiological potential. Seven batches of lentil seeds were subjected to tests to characterise
their physiological potential. The accelerated ageing test included the traditional method (100% RH) with an alternative method
using saturated saline solution (76% RH) at 41 ºC for 24, 36, 48 and 72 hours. NIR spectra were also obtained, taking 200
spectral readings from each batch that were individually processed for 30 seconds. Following the spectral analysis, the seeds
were submitted to tests of germination and accelerated ageing to validate the technique. The PLS-DA technique was employed,
using 70% of the data for training and 30% for validation. Different pre-processing methods were used, including the standard
normal variate (SNV), multiplicative scatter correction (MSC), and the 1st and 2nd Savitzky-Golay (SG) derivatives. It was
concluded that the accelerated ageing test at 41 °C for 48 hours using the traditional method (100% RH) was the most efficient
way of evaluating the physiological potential of the lentil seeds. The models derived from the FT-NIR spectral data were 99%
accurate in predicting the class of physiological potential of the batches.