傅里叶变换近红外光谱技术快速检测金枪鱼新鲜度.docVIP

傅里叶变换近红外光谱技术快速检测金枪鱼新鲜度 　　摘 要：采用傅里叶变换近红外光谱仪和化学计量学方法，探究以近红外为手段的金枪鱼新鲜度指标快速检测技术。选取不同品种、不同贮藏时间下的金枪鱼分割肉块作为研究对象，在4 000～12 000 cm-1波数范围内采集金枪鱼肉样的近红外光谱数据，应用偏最小二乘法（partial least squares regression，PLS）建立K值、硫代巴比妥酸（thiobarbituric acid value，TBA）值、高铁肌红蛋白（metmyoglobin，MetMb）含量、pH值、亮度值（L*）和红度值（a*）等新鲜度指标的定量预测模型，经光谱分析和不同预处理方法优化后，模型能够实现对K值、TBA、MetMb含量、pH值、L*和a*的快速检测，校正相关系数（correlation coefficient of calibration set，RC）分别为0.876 7、0.908 7、0.893 7、0.753 2、0.815 0和0.723 2，内部交互验证校正标准偏差（root mean square error of cross validation，RMSECV）分别为3.641 1、0.654 8、4.700 9、0.053 4、1.286 3和1.317 3。经过外部验证集验证和T检验，模型得到的预测值与真实值之间没有显著性差异（P0.05），说明预测结果准确可靠，可以作为金枪鱼新鲜度检测的有效手段 关键词：近红外；金枪鱼；新鲜度；快速检测 Abstract： The objective of this work was to investigate the potential of Fourier transform near infrared reflectance （FT-NIR） spectroscopy and chemometrics methods to rapidly determine tuna meat freshness. NIR spectra of meat cuts from different tuna species stored for different durations were recorded in the wavenumber range of 4000?12000 cm-1 and preprocessed for the establishment of a quantitative model to predict K value， thiobarbituric acid （TBA） value， metmyoglobin （MetMb） content， pH， L* value and a* value of tuna meat using partial least squares regression （PLSR） method. The results indicated that all above parameters could be predicted quickly and accurately by the developed models with correlation coefficients of calibration set （RC） of 0.876 7， 0.908 7， 0.893 7， 0.753 2， 0.815 0 and 0.723 2， and root mean square errors of cross validation （RMSECV） of 3.641 1， 0.654 8， 4.700 9， 0.053 4， 1.286 3 and 1.317 3， respectively. The models were validated using an external test set and T-test. It was found that no significant difference existed between the predicted values and the actual values （P 0.05）. Accordingly， the models were reliable and could be used to predict the freshness of tuna meat. Key words： near infrared reflectance spectroscopy （NIR）； tuna； freshness； rapid detection DOI：10.7506/rlyj100