Weighted SPXY method for calibration set selection for composition analysis based on near-infrared spectroscopy

An appropriate method for calibration set selection is very important for a quantitative model based on near-infrared spectroscopy. Partial least square regression (PLSR) is one of the most popular regression methods for establishing multivariate calibration models with near-infrared spectroscopic data. However, the success of the PLSR model depends on the availability of a representative set. In this study, we modified the calibration set selection method, Sample set Portioning based on joint x-y distance (SPXY) method, as the weighted SPXY (WSPXY) method. We test the performance of WSPXY method with the published NIR of corn samples data. Then the WSPXY method, and also the Kennard-Stone (KS) method, SPXY method was used to determine the haemoglobin based on NIR spectroscopy. Experimental results showed that WSPXY method can improve the model accuracy. The experimental results verify the performance of the WSPXY method, which can guide the chemical composition analysis based on the spectrum to improve the prediction performance.