基于近红外光谱技术的带鱼新鲜度检测研究

采用傅里叶变换近红外光谱结合偏最小二乘法(PLS)建立了带鱼挥发性盐基氮(TVBN)、pH和菌落总数的定量分析模型。结果表明,TVBN、pH和菌落总数校正集相关系数(r)分别为0.9874、0.9315和0.9990,校正标准差(RMSEC)分别为0.810、0.0961和0.0344,交叉验证均方差(RMSECV)分别为1.81、0.148和0.459。验证集r分别为0.960、0.923和0.994,验证标准差(RMSEP)分别为5.178、0.333和0.823。本研究说明利用近红外光谱技术检测带鱼新鲜度是可行的,为鱼类新鲜度快速检测提供了理论依据。 

Research of detection of ribbonfish freshness based on NIR technique

Quantitative analysis models of ribbonfish TVBN, pH and total numbers of colony were established by Fourier transform near-infrared spectrum (NIR) and Partial Least Squares (PLS) .The results showed that correlation coefficient (r) of calibration set of TVBN, pH and total numbers of colony was respectively 0.9874, 0.9315 and 0.9990.RMSEC was respectively 0.810, 0.0961 and 0.0344.RMSECV was respectively 1.81, 0.148 and 0.459.The r of validation set was respectively 0.960, 0.923 and 0.994.RMSEP was respectively 5.178, 0.333 and 0.823.The research showed that applying NIR to detect ribbonfish freshness was viable, which provided theoretical basis for fast detection of fish freshness.