The Research Of Nondestructive Detection Of Brown Core In Chinese Pear 'Yali' Based On Its Light Physical Properties

Brown core is an internal disorder sometimes seen in pears under storage. The symptoms are not externally recognizable and visible only after cutting the fruit. Because of lacking valid method of nondestructive detection of brown core, the producers and consumers suffer for great losses. It is urgent to find a fast and valid method of nondestructive detection. It is credible and regular to detect the internal quality of fruits based on theirs light physical properties. In this research, Yali's surface color character, transmission light character and reflectance light character were measured and analyzed.CIELAB color variables are convenient method of define color objectively. NIRS technique is applied by more and more researchers because of its characteristics . We classified pears into Browncore and normal pears and grade the Brown core pears to three classes--slight, moderate andsevere.We measured the surface color, NIR spectroscopy and NIT spectroscopy of 236 pears. The data is dealt with SAS and TQ analyst, analysis methods included PLS, PCA and DA. Four different models were established for nondestructive detect of brown core in pears: color discriminance, NIT spectroscopy discriminance, OD difference discriminance and NIR spectroscopy discriminance.From the results of color discriminance, we can learn: When pears were coldly stored for 36 days, most normal pears were incorrectly classified into slight brown. 70% of the moderate brown and 25% of the severe brown were classified into slight brown, too. But none of Brown core were incorrectly classified into normal. With the storage, the error increased.The results of NIT spectroscopy discriminance indicated: When pears were coldly stored for 66 days, none of pears with Brown core were misclassified to normal pears and vice versa. When pears were coldly stored for 108 days, The error of two classes is 2.9%, the error of four classes is 13.3%The classification model of OD difference discriminance was: when its value of A OD(713nm-743nm) was more than -0.231, the pear was classified into brown core; otherwise it was normal pear. According to the model, 5.4% of pears with brown core were misclassified to normal pears and 9.5% of the normal pears were incorrectly classified into brown core.As a result of NIR spectroscopy discriminance, when pears were coldly stored for 108 days, the error of two classes is 3.1%, the error of four classes is 18.8%.Compared with the four method each other, It is distinct that NIT spectroscopy discriminance can predict correctly the pears with brown core, and OD difference discriminance took the second place of prediction. As for color discriminance and NIR spectroscopy discriminance, it is necessary to experiment further for more validly grading.