Egg,as a kind of highly nutritious food,is the daily food products of most people.It has a huge production and sales volume in China,but it is perishable and will endanger human health.Raman spectroscopy is a more widely used and relatively new non-destructive testing and analysis method.By collecting the Raman spectra of the egg surface,it can reflect the information on the surface of the egg shell.This project used Raman spectroscopy technology combined with chemometric methods to establish a quantitative prediction model for five physical and chemical indicators of egg freshness,which provided a new method for the non-destructive,rapid detection and monitoring of egg freshness.The main research contents are as follows:The method conditions for non-destructive testing of egg freshness by Raman spectroscopy was determined.Through comparison,the Raman spectroscopy measurement conditions for eggs were determined as follows: excitation wavelength was 532 nm,integration time wa 2 s,scanning times was 2 times,and detection distance was 8 mm.Under these conditions,samples were measured,and after abnormal data removal,sample diversity,determination of the optimal principal component number,and basic modeling of the original spectrum,the best spectra collection site corresponding to each physical and chemical index was determined.The best site to predict the Haugh unit(HU)and albumen p H(A-p H)was the top of the egg,and the best site to predict the air chamber diameter(ACD)and the air chamber height(ACH)was the bottom.This not only simplified the spectrum collection process,but also reduced errors caused by multi-site collection and improved the representativeness of the collected spectrum.The PLSR quantitative model for non-destructive detection of egg freshness by Raman spectroscopy was established and optimized.A variety of preprocessing methods were used to preprocess the Raman spectroscopy data collected from the best site.The curve smoothing process was suitable for the HU model,the first derivative process was suitable for the yolk index(YI)model,normalization process was suitable for A-p H model,the detrended fluctuation analysis was suitable for ACD model,and original data was suitable for establishing ACH model.The correlation coefficients of the best model calibration set were0.908,0.947,0.938,0.977,0.941,respectively.The correlation coefficients of the prediction set were 0.906,0.862,0.957,0.880,0.862.It indicated that the model had good stability and accuracy,and it proved the feasibility of using Raman spectroscopy technology to non-destructively predict the five physical and chemical indexes of freshness of eggs.On this basis,the spectral data of the main contributing variables in the load diagrams of each model was modeled.It was found that the HU model can realize simplification and optimization.The YI model can be simplified but cannot significantly improved the model performance.The ACH model can also be simplified,but the performance of the model deteriorated.The principles of Raman spectroscopy for non-destructive detecting of egg freshness was explored.It was found that the Raman spectrum of the eggshell surface was related to the organic components of calcium carbonate and the cuticle in the experiment.The comparison of the spectrum showed that the better the quality of the cuticle,the weaker the characteristic peak intensity of calcium carbonate and the stronger the peak intensity of organic components.This method can be used to characterize the quality of egg cuticle.At present,dyeing method was commonly used to evaluate the quality of cuticle.This part first determined that the alternative dye solution for this method was malachite green dye solution,and the dyeing time was 5 min.On this basis,through the analysis of the physical and chemical index values of eggs with different freshness and the quality of the corresponding cuticle,it was found that there was a strong correlation between the two.The pearson correlation coefficient was0.830-0.950.In the meantime,it showed that analyzing the quality of egg cuticle can predict the freshness of eggs.By further comparing the effects of the egg content on the physical form,hydrophobicity and permeability of the cuticle during storage,it was found that the decrease in the freshness of the content would also promote the deterioration of the quality of the surface cuticle,that is,the internal freshness and the quality of the cuticle had a two-way effect.Results again showed that the Raman spectrum information of the cuticle on the egg surface can reflect the internal freshness,and the Raman technology non-destructive detection of egg freshness was theoretically feasible. |