摘要
选取富士苹果和美八苹果的代表性评价指标,进行对比分析建立苹果品种的判别函数,为苹果品种的判别检验奠定基础。分别测定了富士和美八2个品种各300个果实的果形指数、硬度等13项理化指标,进行主成分分析,筛选出其代表性理化指标,进而挑选出5项共有指标为自变量,建立苹果品种判别函数。选取色泽b*值(X_5)、硬度(X_6)、水分(X_7)、VC含量(X_(10))、可滴定酸(X_(11))5项理化指标,运用Fisher判别函数,建立了苹果品种综合判别体系,其判别正确率均达到100%,为苹果品种差异性鉴别提供了一种客观、可行的多指标综合判别方法,为近红外光谱分析技术提供了一种鉴别苹果品种的模型。
The representative evaluation indicators of Fushi and American No.8 fresh apples are selected and analyzed,also the discriminant function of apple varieties is established to lay the foundation for its discrimination. For each variety,thirteen kinds of physical and chemical indicators of 300 fresh apples are determined. Based on principal component analysis method,the representative evaluation indicators are selected. Meanwhile,five communal indicators are picked out as independent variables to establish the discriminant function. b＊（X_5）,hardness（X_6）,water（X_7）,vitamin C content（X_（10））,and titratable acidity（X_（11）） are taken as independent variables and the Fisher discriminant function is used to establish the comprehensive discrimination system of apple varieties,the right rate of which is 100%. This study provides an objective and feasible multi-index comprehensive identification method for the differences identification of apple varieties. And it could provide near-infrared spectroscopy a model which could be used to identify the apple varieties.
出处
《农产品加工》
2016年第2期42-45,共4页
Farm Products Processing
关键词
品种鉴别
主成分分析
Fisher判别函数
综合判别体系
variety identification
principal component analysis
Fisher discriminant function
comprehensive discrimination system
