基于近红外光谱技术的鱼肉新鲜度评价方法的建立

Freshness evaluation of fish meat by near infrared spectroscopy

采用近红外光谱技术和化学计量学方法,测定鳙(Aristichthys nobilis)、鲢(Hypophthalmichthys molitrix)和草鱼(Ctenopharyngodon idellus)等3种鱼肉不同新鲜度样品的挥发性盐基氮(TVB-N)、硫代巴比妥酸(TBA)值、K值和pH值,并同时采集其近红外光谱,以建立基于近红外光谱技术的鱼肉新鲜度的评价方法。结果表明,3种鱼鱼肉样品的新鲜度指标测量值覆盖范围均较大,满足建模要求。根据内部交互验证标准差确定了3种鱼鱼肉新鲜度指标的光谱预处理方法,采用偏最小二乘法(PLS)分别建立了鳙、鲢、草鱼鱼肉新鲜度的近红外定量分析模型。所建模型除鳙鱼肉TVB-N和TBA值模型的预测相关系数(Rp)较小外,其他指标模型的校正相关系数(Rc)和预测相关系数(Rp)均大于0.85,具有较好的拟合度和预测能力,能为鱼肉新鲜度的评价提供一种快速、简便的检测方法。

Total volatile basic nitrogen （TVB-N）, thiobarbituric acid （TBA）value, K value and pH value of meat derived from bighead carp, silver carp, grass carp with different levels of freshness were measured. Meanwhile, the near-infrared spectra were investigated in order to obtain the quantitative analysis methods of fish meat freshness based on near-infrared spectroscopy and ehemometrics methods. The results showed that the freshness measured values of the three kinds of fish meat samples met the modeling requirements. The preprocessing methods of three kinds of fish meat freshness indicators were determined by RMSECV, and partial least squares （PLS） were used to established near-infrared quantitative analysis model for bighead carp, silver carp and grass carp, respectively. The prediction correlation coefficient （Rp） of TVB-N and TBA value of bighead carp meat were small. However, model correction coefficient （Rc） and Rp of other indexes were greater than 0.85, and owed an acceptable fitting accuracy and predictive ability. The results provided a quick and easy detection method for freshness of fish meat.