利用可见近红外光谱检测技术,对南丰蜜橘可溶性固形物含量的无损检测进行了相关研究。在350-1800nm波段,应用偏最小二乘法对经过预处理的光谱进行建模分析。对89个建模样品的建模结果是,建模相关系数和均方差分别为0.955和0.018%;对30个未参与建模样品预测结果是,预测相关系数为0.946,均方差为0.018%。从实验结果可知,可见近红外光谱检测技术对南丰蜜橘可溶性固形物含量的无损检测是可行的。
This research work was to nondestructive inspect the soluble solid content of Nanfeng mandarin by visible/near-infrared spectroscopy. In the wavelength range of 350-1800 nm, Pre-processing methods were used before the spectra were analyzed with partial least square. The results, based on calibration for 89 samples, are 0.943 6 and 0.613 3 for calibration correlation coefficient and root mean square error of calibration. The results, based on prediction for 30 unknown samples, are 0.9304 and 0.6515 for prediction correlation coefficient and root mean square error of prediction, respectively. The results show that visible/near-infrared spectroscopy inspection technology for measuring the soluble solid content of Nanfeng mandarin is feasible.