应用近红外光谱技术快速检测黑加仑浆果的主要营养成分

Rapid detection for main nutrient components of black currant berry us ing near infrared technology

为了快速检测黑加仑浆果中的营养成分,用近红外光谱分析仪测量了黑加仑果浆样品近红外光谱(波长918～1045nm)吸收值,并用多元线性回归(MLR)、主成分回归分析(PCA)、偏最小二乘法(PLS)建立了同时测定黑加仑果浆中总酸、维生素C、总糖、花青素含量的数学模型,并对以上三种回归方法进行统计分析。结果表明,用多元线性回归建立的数学模型更适合预测总酸、维生素C、总糖含量,用主成回归分析建立的数学模型更适合预测花青素含量,对浓度范围分别在1.95%～6.06%,68.6～255.6mg·100g-1,3.13%～8.66%,283.1～669.6mg·100g-1的总酸、维生素C、总糖、花青素含量的预测模型的决定系数(R2)分别为0.9097、0.8826、0.9662、0.9804,校正标准差(RMSEC)分别为0.3324、13.4896、0.1923、14.1208,预测标准差分别为0.1129、2.5210、0.1307、8.4196。结果表明,方法对同时定量分析黑加仑多组分含量是可行的。

A near infrared spectrum method with wavelength range of 918～1045 nm was employed to rapidly detect the absorption value of main nutrient components of black currant berry.Three regression methods viz multiple linear regression （MLR）,principal component analysis （PCA） and partial least square method （PLS） were used to build the regression models to calculate the component content in fresh blackcurrant berry in terms of total acid,vitamin C,total sugar and anthocyanidin.The results of statistical analysis indicated that MLR method is applicable to the detection of total acid （1.95%-6.06%）,vitamin C （68.6-255.6 mg·100 g-1）,total sugar （3.13%-8.66%） and PCA was good for the detection of anthocyanidin （283.1-669.6 mg·100 g-1） with high determination coefficient and low error.Therefore,near infrared technology was suitable to rapidly and simultaneously detect the main nutrients components of black currant berry.