| Near infrared spectroscopy(NIRS)has been widely used in qualitative and quantitative analysis of single grain seeds due to its advantages of fast and nondestructive measurement.This analytical tool overcomes the shortcomings such as time-consuming,large workload and environmental pollution for many other traditional detection methods,and so it may acquire target traits of seeds more accurately and help real-time monitoring of seed quality.Rice is one of the most important food crops in the world.However,few studies have been done on the application of NIRS in rice seed analysis because of the small size and the interference of seed shell.In this study,based on the present research situation,the characteristics of near infrared spectra of single brown rice and rice seed were investigated.And the quantitative models were established successfully,which proved that NIRS could be utilized for the nondestructive quality analysis of single brown rice and rice seed.The major results are summarized as follows:1)The NIRS quantitative calibration model determining moisture property of single brown-rice kernels was established.A partial least squares model selecting three regions,namely,5292-5616 cm-1,7236?7600 cm-1 and 7884?8208 cm-1,with standard normal variate transformation was verified as the best model,which results in a coefficient of determination(R2)of 0.98 and root mean square errors of prediction(RMSEP)of 1.01%.A multivariate linear regression model selecting three wavenumbers,namely,5492.56 cm-1,7158.84 cm-1 and 8285.12 cm-1,was a simple model for producing the optimal results,having a coefficient of R2 of 0.9661 and RMSEP of 1.1137%.2)The near infrared spectral characteristics of brown-rice and rice grains under three measurements(diffuse reflectance,trans-reflectance and transmittance)were investigated,and explored the potential of NIRS to determine protein content in single brown-rice and single rice grains.As a result,in the single brown rice protein content analysis,the quantitative model had better accuracy under trans-reflectance and transmittance spectral measurements,and the most accurate PLS regression was given with external validation set having a coefficient of R2=0.971 and RMSEP=0.267%.In the single rice grains protein content analysis,the quantitative model had a relatively accurate result under the trans-reflectance measurement which had RMSEP=0.806%,and had a very good correlation results under the transmittance measurement which had R2=0.964 and RMSEP=0.244%.These results proves that NIRS can be used as a screening tool in monitoring for the quality of rice without any sample preparation.3)We initiated the investigation of the effect of non-thermal low-temperature plasma on brown rice,recording the changes of brown rice early growth vigor under different plasma treatment conditions,i.e.,different gases,different discharge powers and different treatment time;and furthermore,we attempted to analyze the changes of physicochemical properties for the plasma-treated brown rice based on NIRS for qualitative assessment.The germination test showed that seeds treated with air-plasma had better promotion in growth vigor than the seeds treated with nitrogen-plasma;and so we obtained the optimal treatment conditions,i.e.,360W-5 min treatment for air-plasma,and 360W-10 min treatment for nitrogen-plasma,respectively.The principal component analysis(PCA)of NIR data revealed that there were obvious differences between the treated seeds such as the air-360W-5min group and the nitrogen-360W-10min group and the untreated seeds.This work therefore has demonstrated that the technique of low-temperature plasma can be effectively utilized for promotion of growth of brown rice,while the technique of NIRS is useful for qualitative evaluation of plasma induced effects on rice seeds. |
PDF Full Text Request