Determination of wet gluten in wheat based on wavelet de-noising and PLS

NIRS (Near Infrared Transmittance Spectroscopy) is a new analytic technique in analytical chemistry which is developing very quickly in recent years. It has quick, simple, and nondestructive characteristic. This study is based on the analysis of wheat by near infrared spectroscopy to predict the wet gluten in wheat. Using the wavelet transform to de-noise the spectrum firstly, on the basis of which, the partial least squares model of wheat wet gluten is established. The experimental results indicate that the R, MSE and Er are 0.9711, 1.155 and 3.371% respectively, which certificate that this model could predict the wet gluten in wheat accurately.