Rapid detection of freshness of refrigerated grouper (Epinephelus coioides) fillets by near-infrared spectroscopy and quality index method

Abstract

ABSTRACT The rapid detection of the freshness of grouper fillets was obtained by establishing a quality index method (QIM) scheme and near-infrared analysis model. Between the QI score and storage time showed a linear relationship (QI = 1.179 × t –1.157, R2 = 0.98) which indicates that shelf life of grouper fillets is 12 days under 4 ℃. Partial least squares (PLS) analysis showed the mean squared error between predict days and measure days was almost 1 day (MSE=0.984). Correlation analysis between QI value and freshness indices found that the QI score has a high correlation with total volatile base nitrogen (TVB-N). Partial least squares (PLS), principal component regression (PCR) and multiple linear regression (MLR) methods were used to establish near-infrared spectroscopy (NIRs) prediction models for TVB-N, various spectral pretreatment methods such as the first derivative (1st), vector normalization (SNV), and multi-scatter correction (MSC) have been adopted. The results showed that SNV combined with PLS had the best acceptable fitting accuracy and predictive ability, the coefficients of prediction (Rp) was 0.968 and root mean square error of prediction (RMSEP) was 1.381 for TVB-N. The total results reveal that the feasibility of using NIRS and QIM scheme to detect freshness in grouper fillets.