粮油包括谷类、豆类等粮食作物、油料及其半成品和加工成品,是人类生活的必需品。我国是粮油的生产和消费大国,粮油为人类健康提供了许多营养和功能成分,如淀粉、蛋白质、脂肪酸、氨基酸、维生素、植物甾醇和多酚等[1]。粮油安全关系到国计民生,提高粮油品质、保证粮油安全离不开相关检测技术的发展。粮油的理化指标如色泽、气味、营养成分、过氧化值、酸价等可以用来评价粮油的质量优劣、品质等级、加工适应性和贮存稳定性等特点,也与其品种和产地环境密切相关。粮油产品的安全主要受到食品添加剂、包装物、农药、重金属等有毒化学品的污染,细菌、真菌等有害微生物及其产生的毒素产物,害虫及虫卵对粮油造成的污染,以及生产加工过程中掺入杂质和有害物质等的影响[2]。
许多传统粮油检测技术可用于粮油的品质安全监测,如感官检验法、湿化学方法、薄层层析法、高效液相色谱法、聚合酶链式反应、酶联免疫吸附等,在过去的几十年里得到了广泛应用。这些技术通常需要专业人员操作,样品制备复杂,涉及使用的化学品可能会对环境造成污染,存在破坏样品、效率相对较低、不适用于在线检测等问题[3]。为适应现代检测方法快速、准确的需要,无损检测技术不断创新,对于在线和快检方面展现出应用优势和潜力。本文综述了目前用于粮油的现代无损检测技术以及对粮油主要品质安全指标检测的不同技术研究进展,总结并比较了各检测技术的优势及局限性,并对粮油无损检测未来的发展趋势进行展望。
1 粮油无损检测技术应用现状
无损检测技术是指在不破坏粮油样品的前提下,通过声、光、电、磁、热等物理或化学手段,获取其内部及表面相关信息的技术。近红外光谱、高光谱成像、荧光光谱、拉曼光谱、太赫兹光谱、机器视觉、电子鼻、比色传感器阵列、核磁共振等无损检测技术,在粮油的品质安全检测方面发挥着重要作用(图1)。以下将详细介绍这些技术在粮油无损检测方面的具体应用。
图1 无损检测技术在粮油品质安全检测中的主要应用
Fig.1 Main applications of nondestructive testing technologies in grain and oil quality and safety detection
1.1 近红外光谱(near-infrared spectroscopy,NIRS)技术
NIRS技术是一种基于分子振动光谱的分析方法,它利用近红外光与物质分子作用产生的吸收、散射和反射等现象,对样品的成分和结构进行定量或定性分析[4]。NIRS技术可对粮食中营养成分指标进行评价,在粮食的品种分类、产地溯源、不完善粒筛选、虫害检测、转基因识别等方面有广泛的研究,具体应用如表1所示。此外,SHI等[5]采用NIRS建立了小麦粉中滑石粉和过氧化苯甲酰2种非法添加剂的PLS预测模型,校正决定系数
分别为0.997和0.947,预测决定系数
也达到0.995和0.923;ZHU等[6]采集了不同霉变程度小麦籽粒的傅里叶变换近红外光谱,建立了玉米赤霉烯酮含量的PLS定量检测模型,
为0.954 5,证明了NIRS技术应用于粮食中非法添加剂及霉菌污染检测的可行性。
表1 NIRS技术在粮油检测中的应用
Table 1 Application of NIRS technology in grain and oil detection
检测样本检测指标光谱范围预处理方法特征提取方法建模分析方法模型准确度参考文献燕麦淀粉、蛋白质、β-葡聚糖、脂肪1 350～2 500 nmSNV, MSC, NOR, DT, 1st Der, 2nd Der, BC, SGCARS, SPA, UVE, LARPLSR2P=0.768, 0.853,0.759, 0.903[9]大米品种;掺假1 100～2 498 nmSNV, MSC—PLS-DA96.80%[10]高粱虫害10 000～4 000 cm-1MSC, SNV, SG,1st Der, 2nd Der—PCA, PLS-DA100%[11]小麦不完善粒1 100～2 300 nmSNV, 1st Der, SNV+1st Der—PLS-DA, SAE92.52%[12]玉米转基因400～2 500 nm——PCA, ANN100%[13]小麦粉非法添加剂(滑石粉、过氧化苯甲酰)680～2 600 nmMSC, SNV, DT,1st Der, 2nd DerCARS, RF, MCUVE, CARS-RF,CARS-MCUVEPLSR2P=0.995, 0.923[5]小麦真菌毒素(玉米赤霉烯酮)10 000～4 000 cm-1MSC, SNV, SG CARS, SVM-RFEPLSR2P=0.954 5,RMSEP=18.644 2[6]橄榄油酚类化合物671.82～2 702.7 nmSNV, DT, 1st Der,2nd Der—MPLSR2C=0.84, 0.85,0.88,0.89[14]花生油真菌毒素(黄曲霉毒素B1)12 000～4 000 cm-1SG, 1st Der,2nd Der—PCA, PLS-DAR2=0.951, RMSEC=3.87%,RPD= 4.52[15]花生油品种;掺假950～1 800 nm1st Der, 2nd Der,SG, NOR—PLSR2P=0.997,RMSEP=0.014[7]菜籽油氧化程度12 500～4 000 cm-1SG, 2nd Der—PLSRR2:0.994～0.998[8]花生油重金属(镉)955.773～1 702.646 nmSNV, MSC, SG平滑,1st Der, 2nd DerVISSA, CARS,MFE-LASSO, BOSSPLSR2P=0.966 6,RMSEP=2.820 7 [16]玉米油农药(毒死蜱)10 000～4 000 cm-1——1D-CNNR2P=0.987 4[17]
注:SNV:standard normal variate,标准正态变量变换;MSC:multiplicative scatter correction,多元散射校正;NOR:normalization,归一化;DT:detrend,去趋势化;BC:baseline correction,基线校正;SG:Savitzky-Golay smoothing,SG平滑;1st Der:first derivatives,一阶求导;2nd Der:second derivatives,二阶求导;CARS:competitive adaptive reweighted sampling,竞争性自适应重加权采样;SPA:successive projections algorithm,连续投影算法;UVE:uninformative variable elimination,无信息变量消除;LAR:least angle regression,最小角回归;RF:random frog,随机蛙跳;MCUVE:monte carlo unin formative variable elimination,蒙特卡洛无信息变量消除;SVM-RFE:support vector machine-recursive feature elimination,支持向量机递归特征消除;VISSA:variable iterative space shrinkage approach,变量迭代空间收缩算法;MFE-LASSO:multiple feature spaces ensemble strategy with least absolute shrinkage and selection operator,多特征提取-最小绝对收缩和选择算子;BOSS:bootstrap soft shrinkage,最佳正交子集选择;PLS:partial least squares,偏最小二乘法;PLS-DA:partial least squares-discriminate analysis,偏最小二乘回归;PCA:principal component analysis,主成分分析;SAE:stacked autoencoder,堆栈式自编码器;ANN:artificial neural network,人工神经网络;MPLS:multiway partial least squares,多向偏最小二乘法;1D-CNN:one-dimensional convolutional neural network,一维卷积神经网络;
coefficient of prediction set,预测决定系数;
coefficient of calibration set,校正决定系数;R2:coefficient of determination,决定系数;RMSEP:root mean square error of prediction,预测均方根误差;RPD:relative prediction deviation,相对百分比差异;“—”表示文献未提及(表2、表3同)。
NIRS技术应用于含油量、总黄酮和酚类等成分的定量预测,还能够实现对真菌毒素污染、农药及重金属残留的分析,对油脂的掺假现象、品种鉴别、氧化程度也有一定的应用潜力。孙超仁等[7]将NIRS结合气相色谱快速检测了花生油中的5种脂肪酸,并依据棕榈酸的含量变化来判断掺假现象,最佳模型的相关系数为0.997。OTOKI等[8]将NIRS与液相色谱-质谱法联用,定量分析菜籽油中三酰基甘油氢过氧化氢异构体的浓度,探讨了其氧化原因为自由基和/或光氧化。目前,已在实际生产中应用近红外谷物分析仪快速检测谷物蛋白质、水分、油脂、灰分等关键指标。
NIRS技术具有分析速度快、操作简单、可实现多指标同步检测的优势。但是NIRS技术对样品均匀性要求高,仅聚焦于内部成分的检测,受环境和仪器等因素影响较大,高性能仪器及建模成本较高,对微量元素的检测灵敏度较低等,需要不断提高模型的普适性与稳定性,探索与其他光谱技术、成像技术的有效结合,进一步扩大其应用潜力。
1.2 高光谱成像(hyperspectral imaging,HSI)技术
HSI技术通过采集一系列连续波段的图像形成三维图像数据,融合了传统光谱分析和图像处理技术,能够同步获取目标的空间分布信息和光谱信息[18]。HSI技术在粮油组分含量测定、不完善粒筛选、品种分类、真菌和虫害污染等方面已取得一定成果,具体应用见表2。此外,SUN等[19]采用近红外HSI结合XGBoost模型,融合光谱特征和纹理特征,成功区分不同储藏年份的大米,准确率达98.89%。LI等[20]使用近红外HSI精准区分了国产与进口大豆,并对国产大豆中掺入的不同比例的进口大豆实现了掺假识别。NADIMI等[21]在可见光和近红外波段下利用HSI对亚麻籽种子的机械损伤程度进行了准确分类,并建立了预测种子萌发率的PLS模型,展现了其在种子质量评估中的应用潜力。HE等[22]应用HSI快速测定了小麦粉中的滑石粉,构建了SNV-CARS-PLS最佳预测模型,为0.98。
表2 HSI技术在粮油检测中的应用
Table 2 Application of HSI technology in grain and oil detection
检测样本检测指标光谱范围/nm预处理方法特征提取方法建模分析方法模型准确度参考文献小麦营养成分400～1 000,900～1 700 Pauta criterion, MCCV,1st Der, 2nd Der—PLSR, LASSO, SLRR2>0.6[26]亚麻籽品种380～1 018,870～1 709SG SPA, CARSELM, BPNN, LSTM,1D-CNN95.26%[27]大米储存年份900～1 700 MSC, SNV,1st DerCARS, LASSOXGBoost98.89%[19]大豆产地;掺假884.6～1 728.5SNV, SG UVEPLS-DA100%[20]亚麻籽种子质量(机械损伤、萌发率)397.66～1 003.81,953.36～2 567.37SNV, SG, 1st Der —PCA, PLS90.7%, 85%;R2=0.78, 0.82[21]小麦粉非法添加剂(滑石粉)900～1 700SNV, BC, MSC, GFSSPA, CARSPLSR2P=0.98, RMSEP=2.88,RPD=5.09[22]玉米真菌毒素(黄曲霉毒素和伏马菌素)419～1 007,1 007～2 472NOR, SNV, MSC, SG,1st Der, 2nd Der—SVM, PLS-DA95.70%[28]大豆虫害383.7～1 032.7SG—3D-CNN86%[29]茶油籽含油量374.98～1 038.79SNV, SNV+DT, NOR,1st Der, 2nd DerSPA, GAPLSR, CNN,ACNNR2P=0.829, RMSEP=2.462,RPD=2.425[30]油菜籽成熟度400～1 000SNV, SG, 1st Der,2nd Der, DTCARS, SPA, IVISSAELM, KNN,RF, PLS-DA, SVM97.86%[23]茶籽油氧化程度870～1 720 SNV, SG, NOR,1st Der, 2nd DerSPA, CARSPCR, PLSRR2C=0.969 8, R2P=0.958 1[24]红花籽油掺假400～1 000L2 NN, MSC, MF—RC, LightGBM,RF, GBDT,CatBoost, PLSR2=0.976[25]
注:Pauta Criterion:帕乌塔准则;MCCV:monte carlo cross validation,蒙特卡洛交叉验证;GFS:gaussian filter smoothing,高斯平滑滤波;L2 NN:L2 norm normalization,L2范数归一化;MF:median filtering,中值滤波;GA:genetic algorithm,遗传算法;IVISSA:interval variable iterative space shrinkage approach,改进的变量迭代空间收缩算法;LASSO:least absolute shrinkage and selection operator,最小绝对收缩和选择算子;SLR:stepwise linear regression,逐步线性回归;ELM:extreme learning machine,极限学习机;BPNN:back propagation neural network,反向传播神经网络;LSTM:long short-term memory,长短期记忆网络;XGBoost:extreme gradient boosting,极端梯度提升;3D-CNN:three-dimensional convolutional neural network,三维卷积神经网络;ACNN:attention-based convolutional neural network,注意力机制卷积神经网络;KNN:k-nearest neighbor,K邻近算法;RC:ridge regression,岭回归;LightGBM:light gradient boosting machine,轻量级梯度提升机;GBDT:gradient boosting decision trees,梯度提升决策树;CatBoost:categorical boosting,决策提升。
FENG等[23]通过建立不同机器学习模型对3种不同成熟度的油菜籽实现分类,准确率高达97.86%。GU等[24]融合光谱特征和图像特征建立了茶籽油酸值的定量预测模型,为评估油脂氧化程度提供了判断依据。ZOU等[25]将HSI与气相色谱-质谱联用,构建了红花籽油中掺杂物浓度的预测模型,为食用油的掺假鉴定提供了理论支持。
HSI技术在粮油品质安全检测中具有适合复杂样品检测、高分辨率、高精度、可视化等优势,但其设备比较昂贵,数据采集及处理较为复杂,数据量大,实时在线检测难度较大;模型易受样品、仪器、环境等因素影响,存在通用性差等问题。可通过高效的特征光谱提取与选择方法提升检测速度,研发集特征光谱的采集、处理、指标检测于一体的便携式、在线检测设备,推动基于特征波长的多光谱技术在粮油无损检测中的实际应用。
1.3 荧光光谱(fluorescence spectroscopy,FS)技术
FS技术基于特定荧光物质受激后发射特征波长光的特性,通过分析激发-发射矩阵或特定扫描模式的光谱,实现对目标物的识别与定量检测。FS技术常与PCA、平行因子分析模型和因子判别分析等强大的多元分析工具相结合,对复杂光谱进行解析,实现对粮油产品品质相关安全指标的检测[31]。
SAITO等[32]利用同步FS预测大豆中的蛋白质和油脂含量,模型预测精度分别达到0.86和0.74。LENHARDT等[33]采用三维FS结合平行因子分析模型,基于荧光团相对浓度的差异,对多种谷物粉的分类准确率超过90%。HU等[34]结合弱选择性荧光探针收集大米的激发-发射矩阵光谱,实现了对大米的产地溯源,准确率达100%。KOGNIWALI-GREDIBERT等[35]将前表面FS技术结合PLS-DA模型,有效区分了掺有不同比例木薯粉的小麦粉样品,准确率达96.77%。MATVEEVA等[36]对感染禾谷镰刀菌和链格孢菌早期的小麦籽粒进行检测,分析了其FS之间的差异,为早期真菌病害检测提供了理论依据。
FS技术在油脂检测方面的应用主要包括对脂肪酸、总生育酚、维生素等物质的含量进行定量预测,通过评估油脂的过氧化值、酸值来反应其新鲜度和氧化特性。同时在检测黄曲霉毒素,多菌灵、咪唑胺、百菌清等农药,砷、硒和汞等重金属及多环芳烃等有害物中有广泛的应用,以保障油脂的品质安全。基于在特定发射波长下不同物质的荧光峰变化,对食用油实现品种分类,掺假现象鉴别[37]。表3中总结了该技术在粮油检测方面的具体应用。
表3 FS技术在粮油检测中的应用
Table 3 Application of FS technology in grain and oil detection
检测样本检测指标荧光光谱技术荧光团建模分析方法模型准确度参考文献大豆蛋白质、脂肪EEM, SDSF芳香族氨基酸、黄酮类化合物PLSR, SVMR2p=0.86, 0.74[32]谷物粉品种EEM氨基酸、生育酚、吡哆醇、4-氨基苯甲酸PLS-DA>90%[33]大米产地EEM—M-PCA,U-PLS-DA100%, 98%[34]小麦粉掺假FFS还原型辅酶Ⅰ、维生素B2PCA,CA,PCA-DA,PLS-DA96.77%[35]小麦霉菌(禾谷镰刀菌、链格孢菌)EEM酚类、醌类化合物PCA—[36]橄榄油品种;掺假FFS酚类化合物、维生素EPLSR, SVMR2=0.99, RMSE=0.001 4[38]菜籽油总酚EEM—PLS-DA,PLS, NPLSR2=0.951,RPD= 4.0[39]棕榈油、山茶油、葵花油、紫苏油氧化程度SFS氢过氧化物、醛类、羰基化合物PCAR2=0.973, 0.956,0.970, 0.938[40]菜籽油、花生油农药(多菌灵、咪唑胺)EEM——R=0.994[41]棕榈油重金属(砷、硒、汞)AFS———[42]
注:EEM:excitation emission matrix,三维荧光光谱;SDSF:second derivative synchronous fluorescence,二阶导数同步荧光光谱;FFS:front-face fluorescence spectroscopy,前表面荧光光谱;SFS:synchronous fluorescence spectra,同步荧光光谱;AFS:atomic fluorescence spectrometry,原子荧光光谱;M-PCA:multi-dimensional principal component analysis,多元主成分分析;U-PLS-DA:unfold partial least squares discriminant analysis,无监督偏最小二乘判别分析;CA:clustering analysis,聚类分析;PCA-DA:principal component analysis-discriminant analysis,主成分分析-判别分析;NPLS:N-way partial least squares,N向偏最小二乘法;MLR:multiple linear regression,多元线性回归。
FS技术可以在短时间内检测多个组分,对微量荧光物质的检测具有高灵敏度和特异性,但仅适用于本身具有荧光或可被荧光探针标记的物质,受限于不同样品的荧光特性及样品中的杂质、背景荧光或基质效应等干扰因素。可以通过复杂的样品前处理来减少干扰,开发新型高选择性荧光探针,优化样品抗干扰算法,拓展FS技术在粮油品质安全检测中的应用。
1.4 拉曼光谱(Raman spectroscopy,RS)技术
RS技术是基于激光诱导分子振动作用,通过测量散射光的频移和强度变化,提供样品的分子结构、化学键和官能团等信息[43]。RS技术在粮油微量营养组分、毒素和污染物检测等方面具有一定优势,表面增强拉曼光谱(surface-enhanced Raman spectroscopy,SERS)技术能够提高对毒素及污染物(如玉米赤霉烯酮、黄曲霉毒素B1,多环芳烃、农药)的检测灵敏度。LOGAN等[44]将手持式SERS与金纳米颗粒基底结合,提高了对印度香米中多种农药残留的检测灵敏度,最大残留限量低于10 ppb。WENG等[45]选择甲氧基聚乙二醇-巯基涂层的金纳米棒作为SERS底物,在1 800～400 cm-1范围内检测大米中的有机磷农药残留,最佳模型R2为0.996 7。ZHAO等[46]采用纳米银SERS基底,实现了对染色黑米中痕量苏丹黑B的定量检测,检测限低至0.1 mg/kg。
RS技术可用于评估油料及油脂中的脂肪酸组成、类胡萝卜素含量,以及过氧化值、碘值等稳定性指标,但灵敏度低于NIRS技术,对油脂中的微量成分(如酚类化合物、角鲨烷)有一定的检测优势[37]。ZHAO等[47]运用了多种机器学习算法,成功鉴别了10种食用油品种,最高准确率达96.7%,并基于特征峰变化建立线性回归模型预测食用油掺假程度,最佳模型的R2为0.984。
RS技术具有的高特异性能够提供分子结构信息,适用于多种形态样品,SERS对污染物具有超高检测灵敏度,但高性能光谱仪成本较高,检测结果易受到荧光干扰,样品不均匀可能导致光谱信号不稳定,需提高检测的稳定性和重现性,开发便携、手持式设备,推动新型拉曼光谱技术的应用。
1.5 太赫兹(terahertz,THz)光谱技术
THz光谱技术是利用介于红外和微波之间的电磁波,通过探测物质分子的振动、能级跃迁及分子之间的相互作用,实现对样品内部结构和成分的无损表征[48]。THz光谱技术具有较强的穿透性,对水分含量和状态变化高度敏感,在粮油水分含量迁移、不完善粒识别、病虫害检测、品种溯源、污染物检测、转基因鉴定等方面展现了应用潜力。
JIANG等[49]分析了花生种子的THz图像特异,结合卷积神经网络对霉变、残缺、发芽等不完善粒的识别准确率高达98.7%。YU等[50]将THz成像技术与机器学习相结合,对害虫侵染稻粒的过程进行实时监测,准确率达100%,同时缩短了预测时间。HU等[51]基于苯甲酸在1.94 THz处的吸收峰,建立了小麦粉中苯甲酸添加剂的定量模型,预测相关系数为0.995 6。
在油脂检测方面,LIU等[52-53]研究了特级初榨橄榄油的挥发性有机化合物在脂肪酸组成和吸收光谱上存在的差异,将最小二乘支持向量机与遗传算法相结合进行产地溯源,最优模型的分类准确率为96.25%;将Thz光谱结合BP神经网络和t-SNE特征提取方法对大豆油中的黄曲霉毒素B1进行定量预测,含量低至1 μg/kg的黄曲霉毒素B1的检测准确度超过90%;结合主成分分析-偏最小二乘回归,基于1.174 9～1.503 0 THz特征光谱,预测了花生油在贮藏过程中的过氧化值,预测集相关系数为0.928 9。LIU等[54]采集了转基因和非转基因玉米油的Thz光谱数据,基于PLSR模型对转基因玉米油的识别准确率可达98.7%。
THz光谱技术安全性高、无电离辐射伤害,可穿透包装或表层物质探测内部信息,但存在光谱仪成本极高,易受水分子信号干扰,对低浓度污染物检测灵敏度有限等问题,且在粮油领域缺乏系统化、规模化的研究。未来需重点突破设备成本高的问题,创新抗干扰和散射校正算法及信号增强技术。
1.6 机器视觉
机器视觉基于图像获取、处理、分析技术,结合人工智能算法,实现对粮油外观特征(如颜色、形状、尺寸、纹理、缺陷等)的检测评估,常与光谱技术、电子鼻技术等联用[55]。目前机器视觉技术在粮油方面的应用主要有不完善粒检测、品种分选、虫害、霉变、食用油质量评价等。
主要研究包括基于深度卷积神经网络对玉米种子的品种进行智能分选[56],基于YOLO-V5卷积神经网络模型对大米霉变区域进行计算,对霉变区域检测准率达100%[57],基于多尺度检测器和金字塔网络对常见贮粮害虫实现高精度检测[58],还有将机器视觉技术与NIRS相结合实现了对小麦粉中荧光增白剂含量及全麦粉中脱氧雪腐镰刀菌烯醇的污染水平的快速检测[59-60]。刘士坤[61]研制了基于机器视觉的机收大豆在线破碎率、含杂率检测方法与装置,实现了收获机作业时大豆品质的在线检测。
机器视觉技术主要通过颜色、透明度、沉淀物等特征对油脂质量进行评价。GILA等[62]基于机器视觉技术对初榨橄榄油中水分和不溶性杂质的含量进行了测定。SANAEIFAR等[63]将介电光谱和计算机视觉相融合对橄榄油在贮存过程中颜色、水分、过氧化值、叶绿素和胡萝卜素含量等质量指标的变化进行了表征,实现了在贮存过程中对橄榄油氧化程度和质量特征的监测。
机器视觉技术具有超高的检测效率和速度,适合在线快速分选,但仅限于样品表面信息,对内部成分、内部结构缺陷及非可见污染物等的检测能力有限,且受外界光照等环境和硬件条件影响较大,分析算法适用性有限,对复杂特征的精准识别存在挑战。需要与NIRS、HSI等技术相结合以提高检测精度和范围,开发泛化能力更强的深度学习模型。
1.7 电子鼻(electronic nose,E-nose)技术
粮油中的挥发性有机化合物(volatile organic compounds,VOCs)存在与其质量密切相关的特征信息,E-nose技术通过模拟嗅觉系统,利用传感器阵列捕获样品中释放的VOCs指纹信息,产生信号变化,经放大处理后通过识别算法分析,实现对气味特征的检测和评估[64]。随着气敏传感器的不断发展与更新,E-nose技术在粮油领域的研究主要聚焦于基于气味指纹的品质变化检测(如霉变、虫害、氧化、酸败),应用于粮食的霉变、虫害检测、贮藏加工过程中的品质变化以及对食用油的分类、掺假、新鲜度评估等方面。
CAMARDO LEGGIERI等[65]对玉米中的黄曲霉毒素B1和伏马毒素进行了污染分析,并结合人工神经网络实现了准确的霉变检测。HOU等[66]利用传感器阵列优化的E-nose系统对贮藏小麦中的VOCs进行分析,结合回归算法对贮粮中的害虫密度进行预测,相关系数达到0.96。LI等[67]整合E-nose、电子舌、气相色谱-质谱技术研究了米糠样品在酸败过程中挥发性成分、苦味和理化性质的变化,发现挤压和高压蒸煮处理能显著抑制酸败发生的关键酶(脂肪酶、脂氧合酶)活性。
在油脂检测方面,HAN等[68]将E-nose与气相色谱离子迁移谱法相结合,运用PCA区分了5种不同种类的食用油,对不同掺假率的红花籽油进行了聚类分析。BURATTI等[69]开发了电子感官(电子鼻、电子舌、电子眼)结合数据融合新方法表征不同级别食用橄榄油的特性,以及判别其在贮藏过程中的新鲜度,为评估橄榄油质量和预测货架期提供了新方法。
E-nose技术检测速度显著快于传统色谱方法,样品前处理自动化程度高,设备成本通常低于高端光谱设备,能够处理一些具有刺激性气味的样品。E-nose气味传感器种类较多,但长期使用易发生漂移或老化,且易受环境温湿度的影响,对极低浓度VOCs的检测灵敏度不高,识别模型泛化能力较差。未来需重点研发抗漂移、高灵敏度的新型传感器,开发鲁棒性更高的识别算法,研发小型化、具有特定程序的专用电子鼻系统。
1.8 比色传感器阵列(colorimetric sensor array,CSA)
CSA是利用多种色敏材料(如pH指示剂、金属卟啉、纳米粒子)与分析物(特别是VOCs)的交叉反应,生成独特的颜色变化图谱,从而识别单一组分或区分复杂混合物[70]。ZAREEF等[3]利用CSA技术捕获VOCs实现了对水稻样品的新鲜度和贮存时间的精准分类。ARSLAN等[71]开发了基于智能手机的CSA系统,结合气相色谱法有效地区分了不同地理来源的水稻品种,使用KNN算法识别率达到100%。ZHAO等[72]将卷积神经网络与CSA、NIRS两种技术相结合,检测小麦中玉米赤霉烯酮的污染。KALANTARI等[73]采用单、双金属银、金纳米颗粒作为传感器元件构建了一种CSA,对小麦粉中残留的三唑类杀虫剂进行良好的鉴别和测定。
在油脂检测方面,ZHANG等[74]建立了一种抗干扰CSA技术,开发新型嗅觉可视化装置,结合NIRS用于玉米油中目标重金属的定性和定量检测,具有较高的检测灵敏度和较低的检测限。HUANG等[75]在疏水膜上固定7种pH指示剂和2种金属卟啉构建CSA,利用VOCs与化学染料产生的颜色变化,成功区分了掺杂有不同比例大豆油和玉米油的特级初榨橄榄油。
CSA技术操作简单、响应迅速、成本低廉,适合现场快速筛查,但主要针对能产生有效VOCs或能与色敏材料直接发生反应的物质,且色敏材料易受温湿度、光照等条件影响,对复杂基质中混合物的区分能力有限。提升色敏材料的可选择性、稳定性,结合深度学习算法,集成智能手机等平台,是拓展CSA性能和应用范围的关键方向。
1.9 核磁共振(nuclear magnetic resonance,NMR)技术
NMR技术利用原子核在磁场中的共振特性,通过分析样品中特定原子核的共振信号、弛豫时间(反映分子运动状态)和信号强度,获取样品的化学和物理信息[76]。近年来,NMR技术主要应用于测定粮油中的水分、脂肪、淀粉等理化指标,通过弛豫特性间接分析粮油的硬度、黏度等质构特性,监测粮油在加工贮藏过程中的品质变化;以及基于不同粮油特有的代谢物谱,有效鉴别粮油种类及掺假现象。KOTYK等[77]利用核磁共振成像技术,建立了一种新的高通量测定完整玉米籽粒含油量的方法。何瑶等[78]结合NMR技术和多变量统计分析方法,通过采集稻花香米与掺假样品的氢核磁共振谱,分析其化合物组成差异,实现了对五常稻花香米和掺假大米的鉴别。NG等[79]采用NMR来区分棕榈油、橄榄油、葵花籽油等油中存在的无机氯和有机氯,并定量分析了其中的有机氯含量,回归系数达到了0.999 5以上。STAREC等[80]将定量NMR与高效液相色谱法相结合,实现了对多特级初榨橄榄油中油酸及醛类物质的定量检测。
NMR技术可提供分子水平信息,对水分和油脂检测灵敏度较高,具有非破坏性、分析速度快、样品处理简便等优点,可实现高通量检测。但是其设备成本显著高于NIRS等技术,易受样品均匀性、温度波动及顺磁性物质干扰导致核磁信号不稳定,谱图解析和数据分析专业性强,对农药残留、重金属等污染物的检测灵敏度不足,需与其他技术相结合提高其准确性,从而加强NMR技术的普及和应用。
2 结论与展望
随着相关技术的快速发展,无损检测在粮油领域展现出广阔的应用前景。NIRS、HSI、FS、RS、THz、机器视觉、E-nose、NMR等技术凭借高效、快速、无损的优势,为粮油品质控制和安全提供了有力支持。然而,这些技术在实际应用中仍受限于设备成本较高、检测灵敏度不足、模型普适性差等因素,限制了其在粮油行业的规模化推广。针对当前技术存在的不足,未来可从以下几个方面进行优化与拓展。
第一,推动新兴检测技术的多技术融合,实现软、硬件协同及优势互补。例如NIRS或HSI能够提供样品的内部成分信息,E-nose提供气味信息,两者结合能更早、更全面地评估粮油的新鲜度、氧化程度、早期霉变等指标。NIRS结合FS技术,快速筛查后对高特异性分子进行确认,实现对掺假物及污染物的种类鉴别。THz结合机器视觉技术,提供内部结构与外部形态信息,对包装粮油产品的检测有一定优势。提高对各项指标的检测精度,满足复杂样品的分析需求。同时,建立统一的检测标准和方法,推动技术的规范化应用,确保检测结果的稳定性和可靠性。第二,加快研发低成本、微型化的机器设备,传感器等高性能核心部件,提升设备在复杂环境条件下的抗干扰能力和检测特异性,开发便携式装备、自动化检测系统、智能在线监测系统,推动其在粮油领域的大规模应用,满足现场快速检测的需求。第三,将高性能的设备与智能化的软件系统深度融合,开发高鲁棒性的模型及自适应学习算法,增强识别精度与稳定性,充分利用大数据和人工智能技术,推动多数据融合发展,从而实现高效与准确的数据分析。通过物联网及云计算平台,实现检测数据的远程存储、实时分析及协同控制,满足更多的应用场景。
[1] POUTANEN K S, KÅRLUND A O, G
MEZ-GALLEGO C, et al.Grains-a major source of sustainable protein for health[J].Nutrition Reviews, 2022, 80(6):1648-1663.
[2] 陈瑞鹏. 粮油储运中几种典型真菌毒素的检测方法研究[D].天津:天津科技大学, 2022.
CHEN R P.Study on detection methods of several typical mycotoxins in grain and oil storage and transportation[D].Tianjin:Tianjin University of Science and Technology, 2022.
[3] ZAREEF M, ARSLAN M, HASSAN M M, et al.Recent advances in assessing qualitative and quantitative aspects of cereals using nondestructive techniques:A review[J].Trends in Food Science &Technology, 2021, 116:815-828.
[4] XU Y X, KONG T Y, MA Y F, et al.Near-infrared spectroscopy:Application in ensuring food quality and safety[J].Analytical Methods, 2025, 17(17):3381-3406.
[5] SHI S J, FENG J H, MA Y Y, et al.Rapid determination of two illegal additives in wheat flour by near-infrared spectroscopy and different key wavelength selection algorithms[J].LWT, 2023, 189:115437.
[6] ZHU J W, CHEN Y, DENG J H, et al.Improve the accuracy of FT-NIR for determination of Zearalenone content in wheat by using the characteristic wavelength optimization algorithm[J].Spectrochimica Acta.Part A, Molecular and Biomolecular Spectroscopy, 2024, 313:124169.
[7] 孙超仁, 王凤玲, 王玉玮, 等.基于特征脂肪酸变化对花生油掺伪快速鉴别方法研究[J].食品与发酵工业, 2023, 49(3):296-300.
SUN C R, WANG F L, WANG Y W, et al.Rapid identification of adulteration in peanut oil based on the changes of characteristic fatty acid[J].Food and Fermentation Industries, 2023, 49(3):296-300.
[8] OTOKI Y, ISHIKAWA D, KATO S, et al.Nondestructive determination of canola oil oxidation causes:A near-infrared spectroscopy coupled with liquid chromatography-mass spectrometry for analyzing triacylglycerol hydroperoxide isomers[J].Food Chemistry, 2025, 467:142143.
[9] LI L L, LI L, GOU G Y, et al.A nondestructive detection method for the Muti-quality attributes of oats using near-infrared spectroscopy[J].Foods, 2024, 13(22):3560.
[10] XIE L H, SHAO G N, SHENG Z H, et al.Rapid identification of fragrant rice using starch flavor compound via NIR spectroscopy coupled with GC-MS and Badh2 genotyping[J].International Journal of Biological Macromolecules, 2024, 281:136547.
[11] SANTOS P M, SIMEONE M L F, PIMENTEL M A G, et al.Non-destructive screening method for detecting the presence of insects in sorghum grains using near infrared spectroscopy and discriminant analysis[J].Microchemical Journal, 2019, 149:104057.
[12] 马洪娟, 冀定磊, 赵殿仁, 等.基于NIRS的小麦不完善粒精确快速评定方法研究[J].中国粮油学报, 2024, 39(12):195-200.
MA H J, JI D L, ZHAO D R, et al.A rapid and accurate identification method for unsound wheat kernels based on near infrared spectroscopy[J].Journal of the Chinese Cereals and Oils Association, 2024, 39(12):195-200.
[13] 雷渊雄, 夏阿林, 黄炜, 等.基于近红外光谱结合化学计量学的转基因大豆产地判别[J].食品与发酵工业, 2022, 48(12):275-280.
LEI Y X, XIA A L, HUANG W, et al.Origin discrimination of transgenic soybean based on near infrared spectroscopy and chemometrics[J].Food and Fermentation Industries, 2022, 48(12):275-280.
[14] LI X, MU
OZ-D
EZ C, MIHO H, et al.Evaluation of phenolics in the analysis of virgin olive oil using near infrared spectroscopy[J].Spectrochimica Acta Part A:Molecular and Biomolecular Spectroscopy, 2025, 326:125262.
[15] YAO W Q, LIU R S, XU Z C, et al.Rapid determination of aflatoxin B1 contamination in peanut oil by Fourier transform near-infrared spectroscopy[J].Journal of Spectroscopy, 2022, 2022(1):9223424.
[16] WANG Z Y, DENG J H, DING Z D, et al.Quantification of heavy metal Cd in peanut oil using near-infrared spectroscopy combined with chemometrics:Analysis and comparison of variable selection methods[J].Infrared Physics &Technology, 2024, 141:105447.
[17] MEI C L, XUE Y C, LI Q H, et al.Deep learning model based on molecular spectra to determine chlorpyrifos residues in corn oil[J].Infrared Physics &Technology, 2024, 140:105402.
[18] GUO Z, ZHANG J, WANG H F, et al.Advancing detection of fungal and mycotoxins contamination in grains and oilseeds:Hyperspectral imaging for enhanced food safety[J].Food Chemistry, 2025, 470:142689.
[19] SUN X R, ZHOU X P, LIU C L, et al.Rapid and nondestructive identification of rice storage year using hyperspectral technology[J].Food Control, 2025, 168:110850.
[20] LI X, WANG D, YU L, et al.Origin traceability and adulteration detection of soybean using near infrared hyperspectral imaging[J].Food Frontiers, 2024, 5(2):237-244.
[21] NADIMI M, DIVYANTH L G, CHAUDHRY M M A, et al.Assessment of mechanical damage and germinability in flaxseeds using hyperspectral imaging[J].Foods, 2024, 13(1):120.
[22] HE H J, CHEN Y, LI G L, et al.Hyperspectral imaging combined with chemometrics for rapid detection of talcum powder adulterated in wheat flour[J].Food Control, 2023, 144:109378.
[23] FENG H, CHEN Y Q, SONG J Y, et al.Maturity classification of rapeseed using hyperspectral image combined with machine learning[J].Plant Phenomics, 2024, 6:0139.
[24] GU Y Q, SHI L F, WU J H, et al.Quantitative prediction of acid value of camellia seed oil based on hyperspectral imaging technology fusing spectral and image features[J].Foods, 2024, 13(20):3249.
[25] ZOU Z Y, WANG Q L, LI M H, et al.Combination of gas chromatography-mass spectrometry and hyperspectral imaging for identification of adulterated Safflower seed oil[J].Journal of Food Composition and Analysis, 2024, 135:106593.
[26] SHI T T, GAO Y, SONG J Y, et al.Using VIS-NIR hyperspectral imaging and deep learning for non-destructive high-throughput quantification and visualization of nutrients in wheat grains[J].Food Chemistry, 2024, 461:140651.
[27] ZHU D Y, HAN J Y, LIU C Z, et al.Vis-NIR and NIR hyperspectral imaging combined with convolutional neural network with attention module for flaxseed varieties identification[J].Journal of Food Composition and Analysis, 2025, 137:106880.
[28] KIM Y K, BAEK I, LEE K M, et al.Rapid detection of single- and co-contaminant aflatoxins and fumonisins in ground maize using hyperspectral imaging techniques[J].Toxins, 2023, 15(7):472.
[29] 桂江生, 何杰, 傅霞萍.基于图像检索的大豆食心虫虫害高光谱检测[J].光谱学与光谱分析, 2022, 42(9):2931-2934.
GUI J S, HE J, FU X P.Hyperspectral detection of soybean heart-eating insect pests based on image retrieval[J].Spectroscopy and Spectral Analysis, 2022, 42(9):2931-2934.
[30] YUAN W D, ZHOU H P, ZHANG C, et al.Prediction of oil content in Camellia oleifera seeds based on deep learning and hyperspectral imaging[J].Industrial Crops and Products, 2024, 222:119662.
[31] DANKOWSKA A.5 advances in fluorescence emission spectroscopy for food authenticity testing[J].Advances in Food Authenticity Testing, 2016:117-145.
[32] SAITO Y, ITAKURA K, KURAMOTO M, et al.Prediction of protein and oil contents in soybeans using fluorescence excitation emission matrix[J].Food Chemistry, 2021, 365:130403.
[33] LENHARDT L, ZEKOVI
I, DRAMIANIN T, et al.Characterization of cereal flours by fluorescence spectroscopy coupled with PARAFAC[J].Food Chemistry, 2017, 229:165-171.
[34] HU L Q, ZHANG Y, JU Y, et al.Rapid identification of rice geographical origin and adulteration by excitation-emission matrix fluorescence spectroscopy combined with chemometrics based on fluorescence probe[J].Food Control, 2023, 146:109547.
[35] KOGNIWALI-GREDIBERT S B C, MBOGNING FEUDJIO W, MBESSE KONGBONGA G Y, et al.Front-face fluorescence spectroscopy combined with chemometrics for the discrimination of wheat flour and cassava flour[J].Journal of Food Composition and Analysis, 2024, 127:105962.
[36] MATVEEVA T A, SARIMOV R M, PERSIDSKAYA O K, et al.Application of fluorescence spectroscopy for early detection of fungal infection of winter wheat grains[J].AgriEngineering, 2024, 6(3):3137-3158.
[37] LI X, LIU W W, XIAO L, et al.The application of emerging technologies for the quality and safety evaluation of oilseeds and edible oils[J].Food Chemistry:X, 2025, 25:102241.
[38] ABAMBA OMWANGE K, AL RIZA D F, SAITO Y, et al.Potential of front face fluorescence spectroscopy and fluorescence imaging in discriminating adulterated extra-virgin olive oil with virgin olive oil[J].Food Control, 2021, 124:107906.
[39] SQUEO G, CAPONIO F, PARADISO V M, et al.Evaluation of total phenolic content in virgin olive oil using fluorescence excitation-emission spectroscopy coupled with chemometrics[J].Journal of the Science of Food and Agriculture, 2019, 99(5):2513-2520.
[40] CAO J, LI C, LIU R, et al.Combined application of fluorescence spectroscopy and chemometrics analysis in oxidative deterioration of edible oils[J].Food Analytical Methods, 2017, 10(3):649-658.
[41] CHEN M L, ZHAO Z M, LAN X F, et al.Determination of carbendazim and metiram pesticides residues in reapeseed and peanut oils by fluorescence spectrophotometry[J].Measurement, 2015, 73:313-317.
[42] SOUZA VALASQUES G, MARIA PINTO DOS SANTOS A, LEVI FRANÇA DA SILVA D, et al.Extraction induced by emulsion breaking for As, Se and Hg determination in crude palm oil by vapor generation-AFS[J].Food Chemistry, 2020, 318:126473.
[43] WU Z H, PU H B, SUN D W.Fingerprinting and tagging detection of mycotoxins in agri-food products by surface-enhanced Raman spectroscopy:Principles and recent applications[J].Trends in Food Science &Technology, 2021, 110:393-404.
[44] LOGAN N, HAUGHEY S A, LIU L, et al.Handheld SERS coupled with QuEChERs for the sensitive analysis of multiple pesticides in basmati rice[J].NPJ Science of Food, 2022, 6:3.
[45] WENG S Z, WANG F, DONG R L, et al.Fast and quantitative analysis of ediphenphos residue in rice using surface-enhanced Raman spectroscopy[J].Journal of Food Science, 2018, 83(4):1179-1185.
[46] ZHAO Y B, YAMAGUCHI Y, LIU C C, et al.Rapid and quantitative detection of trace Sudan black B in dyed black rice by surface-enhanced Raman spectroscopy (SERS)[J].Spectrochimica Acta.Part A, Molecular and Biomolecular Spectroscopy, 2019, 216:202-206.
[47] ZHAO H F, ZHAN Y L, XU Z, et al.The application of machine-learning and Raman spectroscopy for the rapid detection of edible oils type and adulteration[J].Food Chemistry, 2022, 373:131471.
[48] 张红涛, 王龙杰, 谭联, 等.太赫兹光谱技术在食品污染检测中的研究进展[J].光谱学与光谱分析, 2024, 44(8):2120-2126.
ZHANG H T, WANG L J, TAN L, et al.Research progress of terahertz spectroscopy in food contamination detection[J].Spectroscopy and Spectral Analysis, 2024, 44(8):2120-2126.
[49] JIANG W B, WANG J, LIN R Q, et al.Machine learning-based non-destructive terahertz detection of seed quality in peanut[J].Food Chemistry:X, 2024, 23:101675.
[50] YU J X, PU H B, SUN D W.Fast real-time monitor of rice grains infested with Sitophilus oryzae based on terahertz imaging combined with machine learning[J].Food Control, 2025, 176:111290.
[51] HU J, LIU Y D, HE Y, et al.Optimization of quantitative detection model for benzoic acid in wheat flour based on CARS variable selection and THz spectroscopy[J].Journal of Food Measurement and Characterization, 2020, 14(5):2549-2558.
[52] LIU W, ZHAO P G, WU C S, et al.Rapid determination of aflatoxin B1 concentration in soybean oil using terahertz spectroscopy with chemometric methods[J].Food Chemistry, 2019, 293:213-219.
[53] LIU W, LIU C H, YU J J, et al.Discrimination of geographical origin of extra virgin olive oils using terahertz spectroscopy combined with chemometrics[J].Food Chemistry, 2018, 251:86-92.
[54] LIU J J.Terahertz spectroscopy and chemometrics classification of transgenic corn oil from corn edible oil[J].Microwave and Optical Technology Letters, 2017, 59(3):654-658.
[55] VELESACA H O, SU
REZ P L, MIRA R, et al.Computer vision based food grain classification:A comprehensive survey[J].Computers and Electronics in Agriculture, 2021, 187:106287.
[56] JAVANMARDI S, MIRAEI ASHTIANI S H, VERBEEK F J, et al.Computer-vision classification of corn seed varieties using deep convolutional neural network[J].Journal of Stored Products Research, 2021, 92:101800.
[57] SUN K, TANG M D, LI S, et al.Mildew detection in rice grains based on computer vision and the YOLO convolutional neural network[J].Food Science &Nutrition, 2024, 12(2):860-868.
[58] LI J T, ZHOU H L, WANG Z M, et al.Multi-scale detection of stored-grain insects for intelligent monitoring[J].Computers and Electronics in Agriculture, 2020, 168:105114.
[59] GUO X X, HU W, LIU Y, et al.Rapid analysis and quantification of fluorescent brighteners in wheat flour by Tri-step infrared spectroscopy and computer vision technology[J].Journal of Molecular Structure, 2015, 1099:393-398.
[60] ZHANG B, JIANG X S, SHEN F, et al.Rapid screening of DON contamination in whole wheat meals by Vis/NIR spectroscopy and computer vision coupling technology[J].International Journal of Food Science &Technology, 2021, 56(6):2588-2595.
[61] 刘士坤. 基于机器视觉的大豆机收破碎率、含杂率在线检测方法与装置[D].合肥:安徽农业大学, 2022.
LIU S K.On-line detection method and device for machine-harvested soybean crushing rate and impurity content based on machine vision[D].Hefei:Anhui Agricultural University, 2022.
[62] GILA A, BEJAOUI M A, BELTRN G, et al.Rapid method based on computer vision to determine the moisture and insoluble impurities content in virgin olive oils[J].Food Control, 2020, 113:107210.
[63] SANAEIFAR A, JAFARI A, GOLMAKANI M T.Fusion of dielectric spectroscopy and computer vision for quality characterization of olive oil during storage[J].Computers and Electronics in Agriculture, 2018, 145:142-152.
[64] JIA W S, LIANG G, JIANG Z J, et al.Advances in electronic nose development for application to agricultural products[J].Food Analytical Methods, 2019, 12(10):2226-2240.
[65] CAMARDO LEGGIERI M, MAZZONI M, FODIL S, et al.An electronic nose supported by an artificial neural network for the rapid detection of aflatoxin B1 and fumonisins in maize[J].Food Control, 2021, 123:107722.
[66] HOU Y X, XIONG L J, LUO X Z, et al.Detection of pest infestation in stored grain using an electronic nose system optimized for sensor arrays[J].Journal of Food Measurement and Characterization, 2025, 19(1):439-452.
[67] LI Y, GAO C, WANG Y, et al.Analysis of the aroma volatile compounds in different stabilized rice bran during storage[J].Food Chemistry, 2023, 405:134753.
[68] HAN L, CHEN M, LI Y T, et al.Discrimination of different oil types and adulterated safflower seed oil based on electronic nose combined with gas chromatography-ion mobility spectrometry[J].Journal of Food Composition and Analysis, 2022, 114:104804.
[69] BURATTI S, MALEGORI C, BENEDETTI S, et al.E-nose, e-tongue and e-eye for edible olive oil characterization and shelf life assessment:A powerful data fusion approach[J].Talanta, 2018, 182:131-141.
[70] JOHNSON N A N, ADADE S Y S, EKUMAH J N, et al.Advances in mechanisms, designs, and applications of colorimetric sensor arrays for food quality control and authenticity verification[J].Trends in Food Science &Technology, 2025, 160:104999.
[71] ARSLAN M, ZAREEF M, TAHIR H E, et al.Discrimination of rice varieties using smartphone-based colorimetric sensor arrays and gas chromatography techniques[J].Food Chemistry, 2022, 368:130783.
[72] ZHAO Y, DENG J, CHEN Q, et al.Near-infrared spectroscopy based on colorimetric sensor array coupled with convolutional neural network detecting Zearalenone in wheat[J].Food Chemistry:X, 2024,22:101322.
[73] KALANTARI K, FAHIMI-KASHANI N, HORMOZI-NEZHADA M R.Development of a colorimetric sensor array based on monometallic and bimetallic nanoparticles for discrimination of triazole fungicides[J].Analytical and Bioanalytical Chemistry, 2022, 414(18):5297-5308.
[74] ZHANG K X, KWADZOKPUI B A, ADADE S Y S, et al.Quantitative and qualitative detection of target heavy metals using anti-interference colorimetric sensor array combined with near-infrared spectroscopy[J].Food Chemistry, 2024, 459:140305.
[75] HUANG L, WANG M Y, LIU H Y.Identification of adulterated extra virgin olive oil by colorimetric sensor array[J].Food Analytical Methods, 2022, 15(3):647-657.
[76] LI X Y, QU L H, SU X, et al.Utilizing low-field NMR for comprehensive quality evaluation of edible oil and oil product[J].Grain &Oil Science and Technology, 2025, 8(1):43-54.
[77] KOTYK J J, PAGEL M D, DEPPERMANN K L, et al.High-throughput determination of oil content in corn kernels using nuclear magnetic resonance imaging[J].Journal of the American Oil Chemists’ Society, 2005, 82(12):855-862.
[78] 何瑶, 郑彦婕, 邓伶莉, 等.五常稻花香米的～1H-NMR波谱分析及掺假鉴别[J].食品工业科技, 2016, 37(12):80-84;171.
HE Y, ZHENG Y J, DENG L L, et al.1H- NMR Spectroscopy and authenticity identification for Wuchang Daohuaxiang rice[J].Science and Technology of Food Industry, 2016, 37(12):80-84;171.
[79] NG M H, NUZUL AMRI I, CHE RAHMAT C M, et al.Potential of nuclear magnetic resonance for the determination of organochlorine in edible oils[J].Journal of Food Composition and Analysis, 2023, 122:105492.
[80] STAREC M, CALABRETTI A, BERTI F, et al.Oleocanthal quantification using (1)H NMR spectroscopy and polyphenols HPLC analysis of olive oil from the bianchera/belica cultivar[J].Molecules, 2021, 26(1):242.