基于拉曼-紫外可见融合光谱技术的进口橄榄油质量等级可视化快速鉴别方法研究  

Visualized Fast Identification Method of Imported Olive Oil Quality Grade Based on Raman-UV-Visible Fusion Spectroscopy Technology

作　　者：邓晓军 马金鸽 杨巧玲 时逸吟 霍忆慧 古淑青 郭德华 丁涛 于永爱 张峰[6] DENG Xiao-jun;MA Jin-ge;YANG Qiao-ling;SHI Yi-yin;HUO Yi-hui;GU Shu-qing;GUO De-hua;DING Tao;YU Yong-ai;ZHANG Feng(School of Kinesiology,Shanghai University of Sport,Shanghai 200438,China;Technical Center for Animal Plant and Food Inspection and Quarantine,Shanghai Customs,Shanghai 200135,China;School of Environmental and Chemical Engineering,Shanghai University,Shanghai 200444,China;Animal,Plant and Food Testing Center,Nanjing Customs,Nanjing 210001,China;Shanghai Oceanhood Opto-Electronics Tech Co.,Ltd.,Shanghai 201201,China;Chinese Academy of Inspection and Quarantine,Beijing 100176,China)

机构地区：[1]上海体育学院运动科学学院,上海200438 [2]上海海关动植物与食品检验检疫技术中心,上海200135 [3]上海大学环境与化学工程学院,上海200444 [4]南京海关动植物与食品检测中心,江苏南京210001 [5]上海如海光电科技有限公司,上海201201 [6]中国检验检疫科学研究院,北京100176

出　　处：《光谱学与光谱分析》2023年第4期1117-1125,共9页Spectroscopy and Spectral Analysis

基　　金：国家重点研发计划项目(2018YFC1603503);上海市农业领域项目(19391901500);长三角科技合作项目(19395810100);上海市技术标准专项(18DZ2201200)资助。

摘　　要：橄榄油因其高营养等特点,成为植物油中日常消费量逐渐增大的主要品类。橄榄油按照加工工艺分为初榨、精炼和混合等不同质量等级。由于不同等级橄榄油价格差异较大,导致橄榄油市场存在以次充好等问题。同时,涉及等级鉴定的指标繁杂,对应的理化检测方法大部分涉及大型实验室设备,检测成本高、效率低且工作量繁重。我国是橄榄油的主要进口国,采用产品标准中逐项指标确认后判定的模式,无法满足目前急速增长的进口产品快速通关要求。该研究聚焦进口橄榄油在口岸监管现场的快速质量评价需求,开发了多光谱信息同时采集和降维融合成像的方法,将紫外-可见光谱与拉曼光谱进行特征数据融合,构建拉曼-紫外可见2D谱图,通过二维成像进行指纹特征判断,构建特级初榨橄榄油、精炼橄榄油以及果渣油的标准2D融合成像源图,作为等级区分标准对照二维谱,进行橄榄油等级可视化判定;结合空间角度值转化算法对橄榄油进行等级定性评判,通过角度值计算得到特级初榨橄榄油与精炼橄榄油的夹角范围在0.7947~1.0947之间,与油橄榄果渣油其值在1.1570~1.3198之间,而特级初榨橄榄油之间角度值均小于0.1,由此可进行不同橄榄油的等级判定;采用角度决策模型进行橄榄油掺杂样品定量分析。制备不同等级橄榄油的混合样本计算得到特级初榨混合精炼橄榄油、果渣油的模型相关系数r分别为0.9942和0.9910,代入不同样本进行验证,相对误差在-4.48%~2.58%之间。采用拉曼-紫外可见融合光谱结合化学计量学建立二维标准谱图对橄榄油等级可视化判定,并建立初榨橄榄油掺伪检测模型进行橄榄油含量的定量分析,实现口岸食品质量和安全风险信息的多维度、高精度和高准确度直观展示。通过采用进口橄榄油质量等级的快速筛查方法,能有效提高口岸关注风险的监测效率,�Olive oil has become the main category with increasing daily consumption of vegetable oils due to its high nutritional characteristics.According to the processing technology,olive oil is divided into different quality grades such as virgin,refining and blending.Because the prices of different grades of olive oil are quite different,the olive oil market has a certain degree of real attribute problems,such as substandard quality.At the same time,there are complex indicators related to grade identification,and most of the corresponding physical and chemical testing methods involve large-scale laboratory equipment with high testing costs,low efficiency and heavy workload.Our country is a major importer of olive oil,and adopting the model of product standard confirmation and determination of indicators one by one,it cannot meet the rapidly increasing requirements for rapid customs clearance of imported products.This research focuses on the rapid quality assessment requirements of imported olive oil at the port supervision site.It develops a method of simultaneous collection of multi-spectral information and dimensionality reduction fusion imaging,which combines the characteristic data of the visible-ultraviolet spectrum and the Raman spectrum to construct the Raman-Ultraviolet,visible 2Dspectrum.Then the fingerprint feature is judged by two-dimensional imaging.The standard 2Dfusion imaging source map of extra virgin olive oil,refined olive oil and pomace oil is constructed,which is used as the grade discrimination standard to compare the two-dimensional spectrum to determine the olive oil grade visually.Finally,combined with the spatial angle value conversion algorithm,the olive oil grade is qualitatively judged.Through the angle value calculation,the angle between extra virgin olive oil and refined olive oil is between 0.7947and 1.0947,and that of olive-pomace oil is between 1.1570and 1.3198.The angle values between the extra virgin olive oils are all less than 0.1,which can be used to determine the grades of differe

关 键 词：融合光谱 可视化 定量分析 橄榄油 质量等级鉴定 

分 类 号：O657.37[理学—分析化学]