Using Near Infrared In Quickly Non-invasave Measurements Of Pear Juice Quality

With the development of beverage industry,drinks adulteration is becoming more and more seriously.In this study,pear juice was taken as the object by the near infrared spectral analysis technology,in order to establish the methods of evaluation and rapid detection of quality of pear.Huangguan pears and xuehua pears were separately Collected from three different growing stages(fresh pears,fruit drop,fruits stored)to each group till 70,after a total of420.Samples were processed through the juicer,each sample was divided into two parts.One part samples had been taken as mearsuring soluble solids(SSC),pH,malic acid,citric acid.Malic acid.Citric acid was measured by high-performance liquid chromatography.Obtaining the Data samples have been used to Calibration and Testing.Another part samples were scanned by near infrared spectroscopy.Spectral scanning range was1100~1650 nm,resolution of 0.2nm.Under such conditions,the original spectrum were obtained.The noise and baseline drift of original near-infrared spectrum could be eliminated by first order derivative,smoothing and different scatter correction.The same method was used to determine different concentrations of pear for correcting and testing data samples.The NIR spectroscopy was obtained by Near infrared which has been through the same spectral processed,all of these are necessary for establishing quality models.Dividing the samples in each group were randomly selected 70 samples of 20 as a validation sample set.The remaining 50 samples were applied as calibration sample set,using principal components regression and partial least squares,modified partial least squares of pear juice forecast.Experiments showed that the MPLS effect best,model calibration coefficient was above 0.85,compared two kinds of pear juice,huangguan pears model correlation coefficient was higher in most cases.Fresh fruit juice model correlation coefficient was the best,fruit drop was the worst.The highest correlation coefficients of fresh crown pear four ingredients cross-validation correlation coefficient were respectively:SSC 0.9546,pH 0.9423,citric acid 0.9401,malic acid 0.8911.The different concentration of model was also obtained the excellent effect.The rapid detection of near-infrared spectroscopy for quality assessment pear could reach to the practical application requirements,with fast and non-destructive character andgood prospects.