基于双波长拉曼光谱的番茄果实贮藏期番茄红素含量无损检测研究

The research of nondestructive quality evaluation technology for fruits and vegetables is of great significance for the development of industry. When using laser-induced Raman spectroscopy to detect lycopene content in intact tomato fruit, the discrimination of lycopene and β-carotene from spectra data is still unsolved due to their severely overlapping. In this study, a local detection method based on dual-wavelength Raman spectroscopy is developed. According to the difference in the absorption spectra of β-carotene and lycopene, lasers with 488 nm and 514 nm wavelength are used to excite the two materials, respectively. The relative content of each material can be obtained by calculating their excitation efficiencies with different wavelength. The research is focused on the following: The correction and reconstruction of Raman spectroscopy are investigated to effectively remove the fluorescence background. Dual-channel Raman probe is designed and realized, and the influences of system parameters and sample traits on measured spectrum are solved. The detection system and method based on dual-wavelength Raman spectroscopy are constructed and verified, and the prediction model of lycopene content in tomato fruit is built. The method and model are applied to monitor lycopene content in tomato fruit during storage nondestructively and dynamically. The results would promisingly provide theoretical basis and technical support for the development of field detection equipment and the control of quality of fruits and vegetables during storage.

果蔬品质无损检测技术的研究对发展现代果蔬产业具有重要意义。在采用激光拉曼光谱对番茄果实中番茄红素进行无损检测过程中，由于番茄红素和β-胡萝卜素的特征峰重叠严重，如何由谱图信息区分二者是至今仍未有效解决的问题。本项目发展了一种双波长拉曼光谱原位检测方法，根据β-胡萝卜素和番茄红素吸收谱带峰位的差异，采用波长为488 nm和514 nm的激光分别激发两种物质，通过计算二者在不同波长下的激发效率，获得二者各自的相对含量。重点研究以下内容：拉曼光谱数据的校正与重建方法，有效去除原始光谱中的荧光背景；双通道拉曼探头的设计与实现，解决系统参数与样品特性对测量光谱的影响；双波长拉曼光谱检测系统和方法的构建与验证，建立番茄果实中番茄红素的预测模型；应用于番茄果实贮藏期番茄红素含量无损动态监测，实现快速无损检测。研究成果有望为现场检测设备研发及果蔬贮藏期品质调控提供理论基础和技术支持。

随着果蔬在贮藏、加工和流通过程中的内部品质监控和快速、无损检测需求的提出，探索如何有效提取和处理果蔬内部的光谱信息，是光谱检测技术能够成功应用于农产品检测领域的关键问题。本项目以番茄为研究对象，发展了基于双波长拉曼光谱技术的番茄红素含量检测新思路。研究优化了拉曼光谱基线校正方法，采用基于非对称最小二乘的平滑器结合二阶导数谱峰检测，能够将拉曼谱线与噪声区分开来，实现准确的谱峰标定及有效的背景去除，获得了良好的番茄果实拉曼信号重建效果；自主设计了488 nm和532 nm双通道拉曼检测探头，基于该探头构建了双波长拉曼光谱检测系统，最大限度消除了两组拉曼信号的系统误差和检测误差，实现了双波长原位表征和分析；在此基础上提出了番茄果实中番茄红素相对含量分析方法，通过β-胡萝卜素和番茄红素在两个波长下的激发效率差异，实现了两种物质的有效区分，从而获得了准确的番茄红素含量，模型相关系数为0.9446，并验证了分析模型的可靠性和稳定性；以该分析方法为技术手段，实现了番茄果实在采后贮藏过程中番茄红素含量的批量、实时、连续动态监测，并研究了采收期（绿熟期、转色期、红熟期）和环境温度（室温22°C、普通低温12°C和冷藏4°C）对番茄果实贮藏过程中番茄红素含量的影响规律。借助于拉曼光谱无损检测技术的优势，番茄红素含量演变的探索得以进一步精细化和准确化，为番茄贮藏期营养成分变化提供了新的认识。研究提出的方法和技术，为拉曼光谱在果蔬品质检测的应用提供了理论基础和技术支持。