Wu Gui-fang, Huang Ling-xia, He Yong
College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310029, China.
Guang Pu Xue Yu Guang Pu Fen Xi. 2008 Sep;28(9):2090-3.
Aiming at the nonlinear correlation characteristic of Vis/NIR spectra and the corresponding sugar content of grape and berries, the Vis/NIR spectra of grape and berries were obtained by diffusion reflectance. A mixed algorithm was presented to predict sugar content of grape and berries. The original spectral data were processed using partial least squares (PLS), and three best principal factors were selected based on the reliabilities. The scores of these 3 principal factors would be taken as the input of the three-layer back-propagation artificial neural network (BP-ANN). Trained with the samples in calibration collection, the BP-ANN predicted the samples in prediction collection. The values of decision coefficient (r2), the root mean squared error of prediction (RMSEP), and bias were used to estimate the mixed model. The observed results using PLS-ANN (r2 = 0.908, RMSEP = 0.112 and Bias = 0.013) were better than those obtained by PLS (r2 = 0.863, RMSEP = 0.171, Bias = 0.024). The result indicted that the detection of internal quality of grape and berries such as sugar content by nondestructive determination method was very feasible and laid a solid foundation for setting up the sugar content forecasting model for grape and berries.
针对葡萄和浆果的可见/近红外光谱与相应糖分含量的非线性相关特性，采用漫反射法获取葡萄和浆果的可见/近红外光谱。提出了一种混合算法来预测葡萄和浆果的糖分含量。利用偏最小二乘法（PLS）对原始光谱数据进行处理，并根据可靠性选择三个最佳主因子。这3个主因子的得分将作为三层反向传播人工神经网络（BP-ANN）的输入。通过校正集中的样本对BP-ANN进行训练，然后对预测集中的样本进行预测。使用决定系数（r2）、预测均方根误差（RMSEP）和偏差值来评估混合模型。PLS-ANN的观测结果（r2 = 0.908，RMSEP = 0.112，偏差 = 0.013）优于PLS的结果（r2 = 0.863，RMSEP = 0.171，偏差 = 0.024）。结果表明，采用无损测定方法检测葡萄和浆果的内部品质如糖分含量是非常可行的，为建立葡萄和浆果的糖分含量预测模型奠定了坚实的基础。
