Noninvasive Quantification of Egg Yolk Ratio Using Vis-NIR Spectroscopy and Multivariate Regression

Abstract

The feasibility of using Vis-NIR spectroscopy to quantify the yolk ratio in light brown shell eggs was investigated. Spectral transmission (600 to 900 nm) data of eggs was used to develop several multivariate regression models to noninvasively predict yolk ratio of eggs. Developed models were then used for prediction of the unknown sample. The best model was found to be Least square support vector machine (LS-SVM) with an accuracy of 88 % and a RMSEP of 0.0408. These results show the potential of Vis-NIR spectroscopy for noninvasive determination yolk ratio in eggs.