Non-Destructive Detection Of Internal Quality Of Salted Eggs During Curing Period Based On Spectral-Imaging

Salted egg is a traditional Chinese egg product,rich in protein,amino acids,iron,calcium and other nutrients,and is favored by consumers.However,the quality of salted eggs is easily affected by factors such as time,temperature and concentration of pickling liquid,and improperly pickled salted eggs not only have poor taste,but also deteriorate in quality,which brings hidden danger to food safety.At present,the quality control of salted eggs and the pickling process mainly rely on manual experience,and the process is tedious and time-consuming as the shells need to be broken and sampled before leaving the factory to check whether the quality is up to standard.Therefore,the egg industry urgently needs relevant technology to realize real-time non-destructive inspection of the internal quality of salted eggs during the curing period.In this paper,fresh duck eggs were cured using a step-by-step curing method,and salted duck eggs of the full curing cycle were used as the research object.The near-infrared diffuse reflectance spectra and visible/near-infrared transmission spectra of salted eggs and the devices for image acquisition were built respectively,and the effective feature parameters characterizing the internal quality of salted eggs were analyzed and extracted by collecting two kinds of spectral and image information of salted eggs.The original spectral data of salted eggs were compared with different pre-processing methods and feature wavelength variable screening methods to establish a spectral-based internal quality detection model of salted eggs;various effective image processing and feature extraction were performed on the transmission images of salted eggs to establish a salted egg yolk index quality prediction model based on image features.Based on the optimal model results,a portable salted egg internal quality detection device was designed.The main research work and results are as follows:(1)To investigate the patterns of changes in quality indexes,correlation analysis and spectral characteristics of salted eggs during the curing period.The changes of internal quality indicators(yolk water content,salted egg yolk index,p H,yolk sodium chloride concentration,egg white sodium chloride concentration,yolk oil content and egg white water content)with the maturity of salted eggs were investigated during the whole curing period of salted eggs.The results showed that both yolk water content and p H tended to decrease gradually until they stabilized after maturation.The trends of sodium chloride concentration of egg yolk,sodium chloride concentration of egg white and salted egg yolk index were the same,and the index values gradually increased until maturity.The oil content of egg yolk and water content of egg white did not change significantly.The correlations among the salted egg quality indexes were high except for the yolk oil content index.In addition,the NIR diffuse reflectance spectrum and visible/near-infrared transmission spectrum were investigated in this paper,and the results showed that the trends of the NIR diffuse reflectance spectrum and visible/near-infrared transmission spectrum were close to the same throughout the curing period of salted eggs.Among them,the NIR spectra at5100 cm-1 to 5200 cm-1 and 6800 cm-1 to 6950 cm-1 wave peaks were correlated with the C-H and O-H multiplication and ensemble frequencies in egg yolk protein and fat.(2)A nondestructive detection model for the internal quality of salted eggs during the pickling period based on spectral information was established.A diffuse reflection and transmission spectral acquisition system was built to collect the relevant spectral data respectively.Through the comparison of spectral pre-processing,spectral feature extraction and modeling analysis,the prediction effect of the model built from transmission spectral data was better than that of the model built from image processing and diffuse reflection spectra.Among the prediction models(PLSR,PCR,MLR and LSSVM)for predicting egg yolk water content,egg white Na Cl concentration,egg yolk Na Cl concentration,p H,and salted egg yolk index,UVE combined with PLSR had the best prediction effect.The correlation coefficients Rp for the prediction set of each model were 0.955,0.9498,0.9218,0.9143,and 0.9208,respectively,and the correlation coefficients Rc for the training set were 0.98,0.9674,0.9332,0.9624,and 0.9464,respectively,with RPDs greater than 2.3.The evaluation index based on the optimal model indicated that the visible/near-infrared transmission spectroscopy acquisition system for non-destructive quality inspection of salted eggs is feasible,which provides technical support for the subsequent development of internal quality inspection devices for salted eggs.(3)A nondestructive detection model of salted egg yolk index based on image features during the pickling period was developed.The salted egg yolk index is an important index to characterize the maturity of salted eggs.The images of salted eggs during the pickling period were collected,and at the early stage of pickling,the yolk area was small,the yolk color was light yellow,and the shell surface was smooth.As the curing time increased,the yolk area became larger,the yolk color gradually deepened to dark yellow,and the shell surface became rough.The salted egg yolk area ratio was obtained using image processing methods such as image enhancement,threshold segmentation and morphological manipulation on the images.The results show that there is a strong correlation between salted egg yolk index,curing days and yolk area ratio.Therefore,a multiple regression model of salted egg yolk index with curing days and yolk area ratio was established in this paper,by which the value of salted egg yolk index can be predicted and the curing maturity condition of salted eggs can be determined,and the overall accuracy R2 of this model reached 0.801,and the F-test statistics were significant,and the model was more reliable.(4)A portable non-destructive testing device for the internal quality of salted eggs and the supporting testing software were developed.The overall design and selection of key components of the nondestructive testing device were described,and the physical construction was carried out.The development of driver calls and GUI pages was studied,the hardware and software were integrated and the model was implanted,and the overall test was carried out.The device was verified and analyzed using salted egg samples,and the applicability of the model was found to be good,and the real-time nondestructive testing of the internal quality of salted eggs could be realized.