论文部分内容阅读
目的利用实验室自行搭建的拉曼点扫描系统,以市售新鲜胡萝卜为研究对象,建立一种快速无损检测胡萝卜中的β-胡萝卜素的预测模型。方法在胡萝卜长度方向每隔1 cm进行点扫描,样品点扫描拉曼光谱曲线均先采用Savizky-Golay 5点平滑法去噪后再用Baseline方法进行去除荧光背景预处理,最后取平均值作为样品的拉曼光谱。胡萝卜中的β-胡萝卜素标准理化值按照国家标准GB/T 5009.83-2003法进行测定,理化值与1521 cm~(-1)和1156 cm~(-1)处β-胡萝卜素的拉曼特征峰强建立了多元线性回归模型。结果模型校正集的相关系数(R_C)和均方根误差(RMSEC)分别为0.9249和12.04 mg/kg,验证集的相关系数(RP)和均方根误差(RMSEP)分别为0.9155和1 1.47 mg/kg。结论基于拉曼光谱完全可以实现新鲜胡萝卜中β-胡萝卜素含量的检测,为新鲜果蔬中β-胡萝卜素含量的定量检测提供技术支撑。
OBJECTIVE To establish a predictive model for fast and non-destructive detection of β-carotene in carrots by using Raman point scanning system built by the laboratory and taking commercially available fresh carrot as the research object. Methods Point-by-point scanning was performed every 1 cm in the length direction of carrot. The Raman spectra were scanned by Savizky-Golay 5-point smoothing method and then Baseline method was used to remove the fluorescence background pretreatment. Finally, the average value was taken as the sample Raman spectrum. The standard physico-chemical values of β-carotene in carrot were determined according to the national standard GB / T 5009.83-2003, the Raman characteristics of β-carotene at 1521 cm -1 and 1156 cm -1 Feng Qiang established a multiple linear regression model. Results The correlation coefficient (R_C) and root mean square error (RMSEC) of the model calibration set were 0.9249 and 12.04 mg / kg, respectively. The correlation coefficient (RP) and root mean square error (RMSEP) of the validation set were 0.9155 and 1.147 mg / kg. Conclusion Based on Raman spectroscopy, the content of β-carotene in fresh carrot can be completely detected, which provides technical support for the quantitative determination of β-carotene in fresh fruits and vegetables.