Nondestructive And Comprehensive Egg Freshness Assessment Based On Spectroscopy Technology

Eggs are highly perishable and will deteriorate during storage.The quality degradation may exert food safety risk and cause adverse effect on consumers and production enterprises.Comprehensive evaluation of egg freshness index and discussion of model maintenance are particularly important for rapid and nondestructive analysis of egg quality.The objects of this research were White Leghorns eggs and Bantam eggs.Based on visible near infrared(VIS-NIR)spectroscopy,orientation of spectral capture,egg freshness single index,egg freshness comprehensive index and model maintenance were sdudied.The contents of this paper are as follow:1.This research was to identify which orientation was better to predict White Leghorns egg freshness based on visible near infrared spectroscopy.The transmission spectra were acquired in the equatorial region and at the blunt end of the eggs.Pretreatment methods including Savitzky-Golay smoothing(SG),multiplicative scatter correction(MSC),the standard normal variate(SNV),the first derivative and the second derivative were used.A partial least squares regression(PLSR)was developed to predict the Haugh unit,yolk index and albumen pH.For the blunt end,the correlation coefficients of the Haugh unit,yolk index and albumen pH were 0.813,0.848 and 0.857,respectively.The best correlation coefficients of prediction sets,gotten from the equatorial region,were 0.881,0.855 and 0.888 for the Haugh unit,yolk index and albumen pH,respectively.The results illustrated that the equatorial region showed a better ability than the blunt end to predict freshness of the White Leghorns eggs.The determination of the better spectral orientation lays a foundation for the subsequent analysis of egg freshness.2.Egg freshness single indexes including egg shape index,yolk index,Haugh unit,albumen pH,air cell diameter and eggshell thickness were studied to build prediction models.Prediction correlation coefficients of Haugh unit were 0.836 and 0.694 for White Leghorns eggs and Bantam eggs,respectively.Prediction correlation coefficients of yolk index were 0.841 and 0.811 for White Leghorns eggs and Bantam eggs,respectively.For air cell diameter,the prediction correlation coefficients of White Leghorns eggs and Bantam eggs were 0.845 and 0.890.The results of albumen pH for two varieties of eggs were the best with Rp of 0.852 and 0.947.The models of egg freshness single index provide data support for prediction of egg freshness.3.The concept of egg comprehensive freshness index(CFI)was proposed.Prediction model for CFI using VIS-NIR spectroscopy was established.For White Leghorns eggs and Bantam eggs,egg shape index and eggshell thickness were not correlated with other indexes,while yolk index,Haugh unit,albumen pH and air cell diameter were significantly correlated with each other.The six indexes were analyzed by the principal component analysis method.Function between the principal component and the single index was obtained by load coefficients.For White Leghorns eggs,the optimal modeling result of CFI obtained by SNV pretreatment(Rc = 0.908,Rp-0.891)was better than the single index.For Bantam eggs,the best modeling results of CFI were obtained by MSC pretreatment,and the correlation coefficients in correction set and prediction set were 0.810 and 0.744,respectively.The establishment of CFI provides a new idea for the comprehensive analysis of egg freshness.4.Model maintenance between White Leghorns eggs and Bantam eggs was discussed to predict CFI and albumen pH.Spectral models were built based on specific samples,and changes of sample could affect model stability and prediction ability.White Leghorns eggs and Bantam eggs are two varieties which vary in egg shape,shell color and intrinsic quality.After test of model compatibility,global updating,direct standardization(DS)and slope/bias correction(SBC)were evaluated to transfer calibration models from one variety to another.For albumen pH,application of the slope/bias correction algorithm obtained the best prediction results.When White Lehorns eggs were as the master variety,the prediction correlation coefficient of albumen pH was 0.89.And the better transfer performance with a Rp of 0,91 was found when Bantam eggs were as the superior master variety.The study of model maintenance provides technical support for future prediction of other varieties,and could expand the application field of spectral model database.5.A nondestructive detection device for egg freshness was constructed.The light source,power supply,spectroscopy and optical fiber were selected by the function of spectral detection.Teflon was selected as white reference.The thickness of white reference was determined by the height of transmission spectra.The position and height of the light source were adjusted by light spot and egg size.The device could predict the egg freshness index by experimental verification and provide hardware support for nondestructive detection of egg freshness.