基于可见光和近红外光谱鲜猪肉蒸煮损失和嫩度检测的研究

蒸煮损失和嫩度是决定猪肉的食用品质的重要指标,文章提出了可见光/近红外漫反射光谱检测真空包装猪肉的蒸煮损失和嫩度的新方法,从而实现对其快速、无损、无污染测定。利用光谱专用分析软件Unscrambler9.6对采集的近红外漫反射光谱分别进行卷积平滑、二阶微分法和多元散射校正预处理,用偏最小二乘法(PLS)建立其定量校正模型。结果表明近红外光谱漫反射法的预测值与常规方法测定值的相关系数分别为0.81和0.78。该研究结果说明基于可见光/近红外光谱漫反射光谱的检测方法简便易行,是无损检测猪肉的蒸煮损失和嫩度的较好方法。     

近红外光谱测定猪眼肌肌内脂肪中脂肪酸含量

为了快速无损测定猪眼肌肌内脂肪中脂肪酸含量,提出了近红外光谱技术检测的新方法。选用气相色谱法作为猪眼肌肌内脂肪中脂肪酸测定的标定方法,采用漫反射光纤探头采集由低密度聚乙烯塑料袋包装的猪眼肌样品的光谱,来测定眼肌内脂肪中脂肪酸的含量。利用光谱专用分析软件Unscrambler9.6, 对采集的光谱用平滑,二阶微分法进行预处理, 用偏最小二乘法(PLS)建立其定量检测模型。用于本实验的样本总数为82,将异常点除去后,将样品分为校正集和检验集。用校正集的样品来建立定标方程,再用它来预测检验集样品的值。结果表明,脂肪酸C14∶0,C15∶1,C16∶0,C16∶1,C18∶0,C18∶1,C18∶2,C18∶3,C20∶1,C20∶4,SFA,MUFA和PUFA校正模型的相关系数分别为0.57,0.76,0.71,0.77,0.62,0.81,0.86,0.91,0.85,0.91,0.67,0.81和0.95。可见, 用该方法测定肌内脂肪中脂肪酸有较高的测定精度,因此,应用近红外光谱对猪眼肌肌内脂肪中脂肪酸含量的快速无损检测具有可行性。     

水果坚实度的近红外光谱检测分析试验研究

应用傅里叶漫反射近红外光谱技术探讨了水果坚实度无损检测的方法。利用偏最小二乘法建立了坚实度与漫反射光谱的无损检测数学模型,同时对不同光谱预处理方法和不同建模波段范围对模型的预测性能进行了对比分析。结果表明：利用傅里叶变换光谱仪采集的原始光谱的平滑预处理对结果并没有太大影响;原始光谱在800～2 500 nm范围的模型得到了最好的预测结果：校正集样本的相关系数r为0.869,校正均方根误差RMSEC为3.88 N;预测集样本的相关系数r为0.840,预测均方根误差RMSEP为4.26 N。 通过本研究得出：应用近红外漫反射光谱检测水果坚实度是可行的,为今后快速无损评价水果成熟度提供了理论依据。     

近红外漫反射光谱检测赣南脐橙可溶性固形物的研究

研究了应用可见-近红外漫反射光谱技术快速检测赣南脐橙可溶性固形物的方法。以40个赣南脐橙为标准样本,利用漫反射光谱测定法获取完整赣南脐橙的可见-近红外光光谱(350～2 500 nm),采用多种光谱校正算法,选取不同的光谱波段范围对水果样本的漫反射二阶光谱进行有效信息的提取和分析,并结合偏最小二乘法和主成分回归等定量校正方法,建立了赣南脐橙可溶性固形物的定量数学模型。实验结果为: 在361～2 488 nm波段范围内,偏最小二乘法校正模型的预测精度最好,校正模型的相关系数为0.929,校正标准偏差和预测标准偏差分别为0.517,0.592,其预测集样本的预测值与真实值的相关系数为0.791。实验结果表明：应用近红外漫反射技术对赣南脐橙可溶性固形物的快速无损检测具有可行性。     

近红外漫反射光谱法测定整粒小麦单株蛋白质含量

应用近红外漫反射光谱技术(波长为1 100～2 498 nm,分辨率为2 nm),以整粒小麦为材料建立适合于小麦单株分析的蛋白质含量分析系统。首先选取籽粒蛋白质含量具有梯度差异的小麦样品,然后对样品扫描得到原始光谱信息,通过散射校正及数学处理来消除原始光谱噪声,最后分别采用多元线性回归、主成分分析法和偏最小二乘法法建立回归方程。结果表明,优化各项参数后,光谱经过标准乘性散射校正和一阶导数处理后,回归分析采用修正的偏最小二乘法(MPLS)得到的定标模型效果最佳。最终得到的预测方程定标相关系数(RSQ)、交叉验证标准误差(SECV)、交叉验证相关系数(1-VR)分别为0.94,0.42,0.87。数学模型经过验证样品集检验,预测相关系数达到0.88。该模型达到了快速、无损分析单株小麦的要求,非常适合于品质育种的早代选择。     

猪肉pH值的可见近红外光谱在线检测研究

pH值是猪肉关键品质之一,实施在线检测对优化肉品加工工艺、保证产品质量、提高肉及肉制品的经济价值有重要意义。研究应用可见近红外光谱对新鲜猪肉pH值进行在线检测,实验时样品以0.25 m·s-1的速度运动,采集其可见近红外漫反射光谱(350～1 000 nm),进行反射距离校正后应用偏最小二乘回归法建立猪肉pH值在线检测模型。研究通过Kennard-stone算法划分样品校正集与预测集,对比了不同的光谱预处理方法(多元散射校正,微分等)对预测结果的影响,并对建模所用光谱变量进行优化。研究发现经过多元散射校正结合一阶微分预处理的模型效果最好,模型预测相关系数为0.905,预测均方根误差为0.051,经过优化的模型建模所用波长变量数减少一半,模型的预测相关系数提高到0.926,预测均方根误差下降至0.045。结果表明可见近红外光谱可用于新鲜猪肉pH值的在线检测。     

利用近红外光谱技术建立土壤水分含量定标方程

应用近红外漫反射光谱技术（波长为1100—2498nm,分辨率为2nm）,以温县和郑州市郊的土壤为材料建立土壤水分含量分析系统。结果表明,优化各项参数后,光谱经过标准正态变换校正和一阶导数处理后,回归分析采用修正的偏最小二乘法（M PLS）得到的定标模型效果最佳。最终得到的预测方程定标相关系数（RSQ）、交叉验证标准误差（SECV）、交叉验证相关系数（1-VR）分别为0.98,0.92,0.97。数学模型经过验证样品集检验,预测相关系数达到0.96,达到了快速、准确分析土壤水分含量的要求。     

棉籽油分含量近红外无损检测分析模型与应用

棉花是一种重要的油料作物。建立快速、无损检测棉花种子含油量的方法,对于棉花油分育种工作中的材料鉴定、筛选具有重要意义。利用近红外光谱仪采集118份不同油分含量棉花种籽的近红外漫反射光谱,结合化学方法测定验证,建立了棉籽油分含量快速无损检测的近红外模型。光谱预处理方法采用一阶导数＋多元散射校正(MSC),光谱范围5 446～8 848 cm-1,主成分维数为5,以基本覆盖陆地棉棉籽含油量范围的106份试验材料为校正样品集,利用偏最小二乘法(PLS)建立了棉籽仁油分含量近红外反射光谱(NIR)校正模型。校正模型决定系数R2=0.975,校正标准差SEC=0.67。用外部验证样品集进行外部验证,对所建模型的实际预测能力进行检验。结果表明,油分含量预测值与化学值相关系数r=0.978,预测结果误差范围0.1%～1.7%,建立的模型具有很好的预测性。利用建立的模型对784份育种材料进行了油分含量预测,结果显示,该模型应用可以加快棉花育种材料的油分鉴定。     

近红外光谱法快速检测猪肉中挥发性盐基氮的含量

为了实现快速无损地检测猪肉新鲜度的目的,应用近红外光谱法测定猪肉新鲜度重要指标一挥发性盐基氮(TVB-N)的含量.猪肉原始光谱经标准偏差归一化方法(SNV)预处理后,用联合区间偏最小二乘法(siPLS)建立猪肉预处理后光谱和TVB-N含量的校正模型并与经典偏最小二乘法(PLS)模型、间隔偏最小二乘法(iPLS)模型作比较.试验结果表明,利用联合区间偏最小二乘法所建的预测模型最佳,其校正集相关系数(Rc)和交瓦验证均方根误差(fRv)分别为0.8332和3.75,预测集的相关系数(Rp)和预测均方根误差(fRP)分别为0.8238和4.17.研究结果表明利用近红外光谱和联合区间偏最小二乘法可以快速地测定猪肉中挥发性盐基氮的含量.     

葡萄浆果糖度可见/近红外光谱检测的研究

针对可见/近红外光谱与水果糖度存在非线性相关的特点,利用漫反射光谱测定方法获取了葡萄浆果的可见/近红外光谱,提出了应用偏最小二乘（PLS）结合人工神经网络（ANN）建立葡萄浆果糖度的预测模型,利用偏最小二乘法（PLS）对原始光谱数据进行处理,得出交叉检验的最佳主因子数为3,并将3个主因子的得分作为三层BP神经网络的输入。通过定标集样本对BP神经网络进行训练,用优化的BP神经网络模型对预测集样本进行预测。PLS-ANN模型对样本的预测模型检验参数r2为0.908,RMSEP为0.112,Bias为0.013,好于只使用PLS模型的预测模型检验参数r2为0.863,RMSEP为0.171, Bias为0.024。结果表明,利用近红外光谱技术无损检测葡萄浆果糖度等内部品质是可行的,为今后进一步分析建立浆果内部品质预测模型奠定了基础。     

基于光谱技术和支持向量机的生鲜猪肉水分含量快速无损检测

为实现生鲜肉水分含量的快速无损检测,在波长350～1 700 nm范围内采集生鲜猪肉98个样本的可见近红外反射光谱。经中值平滑滤波、多元散射校正和一阶微分复合预处理方法对原始光谱进行降噪处理。将样本数据随机分为训练集和测试集,以训练集交叉验证网格搜索法确定最佳惩罚参数,利用径向基核函数的支持向量机算法建立了支持向量机预测模型,并与偏最小二乘回归建模法进行比较。用径向基核函数的支持向量机算法所建模型对生鲜肉水分含量进行预测的结果为：训练集的预测相关系数R_c为0.96、标准差SEC为0.32,测试集的预测相关系数R_v为0.87、标准差SEV为0.67。实验结果证实用支持向量机所建模型适合于生鲜猪肉水分含量的无损快速检测。     

The impact of quick-freezing methods on the quality,moisture distribution and microstructure of prepared ground pork during storage duration

The aim of present study was to investigate the influences of ultrasound-assisted immersion freezing (UIF), immersion freezing (IF) and air freezing (AF) on the quality, moisture distribution and microstructure properties of the prepared ground pork (PGP) during storage duration (0, 15, 30, 45, 60, 75 and 90 days). UIF treatment significantly reduced the freezing time by 60.32% and 39.02%, respectively, compared to IF and AF (P < 0.05). The experimental results of quality evaluation revealed that the L* and b* values, juice loss, cooking loss, TBARS values and carbonyl contents were decreased in the UIF treated samples, while the a* value, peak temperatures (T_m), enthalpy (ΔH) and sulfhydryl contents were significantly higher than those of IF and AF treated samples (P < 0.05). In addition, low-field nuclear magnetic resonance (LF-NMR) and differential scanning calorimetry (DSC) analysis demonstrated that UIF inhibited the mobility of immobilized water and reduced the loss of immobilized and free water, and then a high water holding capacity (WHC) was achieved. Compared to the IF and AF treatments, the UIF treated PGP samples possessed better microstructure. Therefore, UIF could induce the formation of ice crystals with smaller size and more even distribution during freezing process, which contributed to less damage to the muscle tissue and more satisfied product quality.     

基于APTLD的猪肉中莱克多巴胺残留含量的三维同步荧光光谱快速测定

应用三维同步荧光光谱法结合交替惩罚三线性分解(APTLD)来建立猪肉中莱克多巴胺残留含量的定量测定模型,以实现猪肉中莱克多巴胺残留含量的快速测定。首先分析了莱克多巴胺的荧光光谱产生机理和样本的三维同步荧光光谱;其次对猪肉提取液中的莱克多巴胺荧光的浓度猝灭现象进行了分析;然后应用核一致诊断法确定了APTLD的三线性分解组分数为2,并建立了猪肉提取液中莱克多巴胺的相对荧光峰值强度与训练样本中莱克多巴胺的相对荧光峰值强度之间的标定曲线,用于待测样本中的相对荧光峰值强度的校正;最后,建立了基于APTLD的猪肉中莱克多巴胺残留含量的三维同步荧光光谱预测模型。试验结果表明,该方法可以较好的解决猪肉样本中莱克多巴胺与背景之间的同步荧光光谱严重重叠的问题,省去了一些烦琐的“化学分离”过程,模型预测集的决定系数(R2)和均方根误差(RMSEP)分别为0.986 3和0.496 6 mg·L-1,达到了猪肉中莱克多巴胺残留含量快速定量测定目的。     

应用近红外漫反射光谱对猪肉肉糜进行定性定量检测研究

利用傅里叶变换近红外漫反射光谱结合不同数学建模算法对不同部位取样的猪肉肉糜进行定性判别建模,并建立猪肉肉糜品质指标脂肪、蛋白质和水分含量的定量检测模型。结果表明:不同部位取样猪肉肉糜判别分析PLSDA模型性能良好,最优模型校正集判别正确率为100%,预测集判别正确率为96%;比较两种方法结合,不同光谱预处理建立各品质指标的定量模型,LS-SVM模型性能优于PLSR模型,脂肪和水分含量最佳预测模型校正及预测相关系数r均高于0.9,蛋白质含量最优模型校正及预测相关系数r,RMSEC,RMSEP和RMSECV分别为0.722,0.593,1.595,1.550和1.888,模型精度需进一步提高。研究表明利用傅里叶变换近红外漫反射光谱快速判别不同部位猪肉肉糜的方法是可行的,脂肪和水分含量定量分析模型从预测精度、稳定性及适应性考虑均具一定的通用性,具有良好的市场应用前景。     

空间分辨散射光谱多参数信息融合方法的生鲜肉嫩度无损检测

嫩度是猪肉食用品质最重要的指标之一。猪肉嫩度取决于猪肉组织复杂的物理、化学特性,目前难以实现快速无损伤检测。探索空间分辨光谱技术用于生鲜肉嫩度无损检测的可行性。首先利用点光源高光谱扫描系统采集54块猪肉背最长肌的空间可分辨散射光谱,经过感兴趣区域选择,提取出猪肉样本表面光斑的空间扩散轮廓,结合4-参数洛伦兹分布函数对扩散轮廓进行非线性拟合,拟合优度R2>0.992,并通过残差分析,表明4-参数洛伦兹分布函数符合肉样表面光强的空间散射规律,进而提取出480～950 nm波长下空间分辨光谱的四个形态学参数：渐进值a、峰值b、半带宽c以及半带宽处的斜率d。然后将单参数谱分别与猪肉样本Warner-Bratzler剪切力（WBSF）测量值进行偏最小二乘回归（PLSR）分析。结果表明不同参数谱都含有猪肉嫩度信息,其中峰值参数谱b建模效果最佳,其回归模型的校正集决定系数R2_c为0.674,均方根误差SEC为8.396N,预测集决定系数R2_p为0.610,均方根误差SEP为8.643N。为提高模型的预测精度和稳定性,实现多参数谱信息的融合,先通过PLSR分析,分别提取出每个参数谱中对猪肉嫩度方差贡献大的公共因子,然后将其因子得分组合在一起作为参数谱的特征变量,与猪肉样本WBSF测量值作多元统计回归分析。为避免数据冗余,对不同参数谱特征变量进行多重共线性判别,进一步采用PLSR算法对参数谱特征变量进行降维和变换,采用交叉验证方法,选择前两维因子得分进行校正模型的建立。其中所提取第一维公共因子对猪肉WBSF值方差解释率达92.28%。与单参数谱所建PLSR模型相比,多参数谱信息融合模型预测效果有了较大提高,其R2_c和R2_p分别为0.923和0.800,SEC和SEP分别为4.083N和5.655N。通过对回归系数进行统计量t检验,结果表明所有回归系数极显著（p<0.01）。本研究通过采取多参数信息融合方法为空间分辨光谱在生鲜肉嫩度无损检测应用提供一种思路,该方法有效将空间分辨光谱解析为4个形态学参数,并实现不同参数谱信息的提取和融合,为开发基于空间分辨光谱的生鲜肉嫩度无损快速检测装备提供技术支撑。     

Synergistic effects of oat β-glucan combined with ultrasound treatment on gel properties of silver carp surimi

The effect of oat β-glucan (OG) combined with ultrasound treatment on the gelation properties of silver carp surimi with different salt contents was investigated. The results demonstrated that the gelation properties of surimi gels at high salt concentration were superior than those at low salt level. The addition of OG or ultrasound treatment could significantly enhance the texture properties, gel strength and water holding capacity (WHC) of gel samples, regardless of salt contents. The ultrasound treatment improved the whiteness of surimi gels, whereas the OG addition slightly declined the whiteness. Both OG addition and ultrasound treatment markedly reduced the total sulfhydryl content (total SH) and strengthened the hydrophobic interactions, forming the more uniform and denser gel network structures, hence more water was captured in network structures and became immobilized. Moreover, the combined treatment of OG and ultrasound showed synergic action on the gelation properties of surimi, and the gel strength and WHC of low-salt surimi gel treated by the combination of OG and ultrasound were even superior than that of high-salt gel without OG by traditional heating.     

Evaluation of effects of ultrasound-assisted saucing on the quality of chicken gizzards

The purpose of this study was to evaluate the effects of ultrasound-assisted saucing on the quality of chicken gizzards. The results showed that with the prolonging of the saucing time, the yield, water holding capacity (WHC), lightness (L*), redness (a*) and springiness of chicken gizzards significantly decreased, while the shear force, hardness and chewiness significantly increased (P < 0.05). When the saucing time was the same, the yield, WHC, springiness and tenderness of the ultrasound group were significantly higher than those of the non-ultrasound group (P < 0.05). In particular, when the saucing time was 30 min, the yield, WHC and springiness of the ultrasound group increased by 2.13%, 0.97% and 10.53%, and the shear force decreased by 21.22% compared with those of the non-ultrasound group, respectively. Besides, ultrasound pretreatment increased the content of aromatic compounds, short-chain alkanes, alcohols, aldehydes and ketones, and the principal component analysis displayed that C-50 (saucing for 50 min without ultrasound pretreatment) and U-30 (saucing for 30 min with ultrasound pretreatment) were similar in flavor. Therefore, ultrasound pretreatment is a potential way to improve the quality of saucing chicken gizzards and shorten the processing time.     

Covalent binding of ultrasound-treated japonica rice bran protein to catechin: Structural and functional properties of the complex

Due to the existence of many disulfide bonds in japonica rice bran protein (JRBP) molecules, their solubility is poor, which seriously affects other functional properties. To improve the functional characteristics of JRBP molecules, they were processed by ultrasound technology, and JRBP-catechin (CC) covalent complex was prepared. The structural and functional properties of indica and japonica rice bran proteins and their complexes were compared; furthermore, the changes in the structural and functional properties of JRBP-CC under different ultrasound conditions were investigated. The results showed that compared with indica rice bran protein (IRBP), the secondary structure of JRBP-CC was very different, the water holding capacity (WHC) was higher, and the emulsification performance was better. Different ultrasound conditions had different effects on the functional properties of JRBP-CC. When the ultrasound power was 200 W, the λ_max redshift of the JRBP-CC complex was the most significant, the particle size was the smallest, the absolute value of the zeta potential was the largest, and the hydrophobicity and microstructure of the JRBP-CC complex were the best. Concurrently, the maximum WHC and oil holding capacity (OHC) of JRBP-CC under these conditions were 7.54 g/g and 6.87 g/g, respectively. Moreover, the emulsifying activity index (EAI) and emulsifying stability index (ESI) were 210 m²/g and 47.8 min, respectively, and the scavenging activities of 1,1-diphenyl-2-picrylhydrazyl (DPPH) and ABTS⁺ were 71.96 % and 80.07 %, respectively.