Composition Analysis of Milk by Using Near Infrared Diffuse Reflection Spectroscopy

In order to improve the analysis efficiency, this paper studies the method of analyzing the composition of milk by using nearinfrared diffuse reflection spectroscopy and discusses the propagation of the light in the milk and some relevant measuring methods. The model is built by partial least-squares (PLS),pretreatment methods and different spectral regions are compared. The best correlation values of fat and protein are 0.99 and 0.92 respectively, the best root mean square error of prediction (RMSEP) values of fat and protein are 0.048 g/dL and 0.102 g/dL respectively. The study shows that near infrared diffuse reflection spectroscopy is a potentially useful technique for evaluating the composition of milk.