【摘要】 为增强模型的适应性,选取了3个不同成熟期(绿熟、半红熟和红熟)的李果实样品建立坚实度指标的近红外检测模型,建模所使用的光谱范围为4 000~12 492cm-1。为改善模型性能,比较了最小二乘支持向量机和偏最小二乘法两种建模算法对李果实坚实度指标的建模结果。研究结果表明,所建立的最小二乘-支持向量机模型的预测性能和稳定性均好于偏最小二乘模型,并以前10个潜在变量得分作为输入变量的最小二乘-支持向量机模型为最佳模型,其校正相关系数、校正和预测均方根误差分别为0.989及1.31、1.84kg/cm2,剩余预测偏差为4.79。与以往研究文献相比,获得了较为理想的预测精度和稳定性能。研究结果表明,最小二乘支持向量机算法结合偏最小二乘法提取的潜在变量作为输入变量,可以使李果实坚实度近红外定量模型有较大程度的改善。更多还原

【Abstract】 Plum samples of three different maturity stages(green-maturity,prered-maturity and red-maturity stages)were chosen to establish near infrared spectroscopy( NIR) models to quantify firmness of plums for a wider range of models application. The spectral region used was 4000-12492 cm-1. In order to improve models performance,Least squares-support vector machine( LS-SVM) with latent variables( LVs),extracted by partial least squares( PLS),as input were used to establish calibration models. And the performance were compared with PLS models. LS-SVM models were superior to PLS model in calibration,prediction and robustness. Optimal models were obtained by LS-SVM with the first 10 LVs as input. The correlation coefficients of calibration and root mean square error of calibration and prediction were 0. 989,1. 31 kg / cm2 and 1. 84 kg / cm2. The residual predictive deviation was 4. 79,which were more satisfied in prediction accuracy and robustness than results reported by previous works. The results indicate that with LVs as input LS-SVM offers more effective quantitative capability for firmness of plum.更多还原