摘要

Total volatile basic nitrogen (TVB-N) content is one of the core indicators for evaluating freshness of duck meat. Visible and near-infrared (Vis/NIR) reflectance spectroscopy was implemented in this study to determine the TVB-N content in duck breast meat. Quantitative calibration models were built by partial least square regression (PLSR) between the spectral data and the measured TVB-N values. The different spectral pre-processing methods were employed and synergy interval partial least squares and principal component analysis methods were used to select important wavelengths. In comparison, the prediction model with full spectra after multiplicative scatter correction pre-processing yielded optimum results with a root mean squared error for the prediction set (RMSEP) of 1.060 mg/100 g and a correlation coefficient (RP) of 0.859. The results of this study demonstrated the feasibility of the quantification method for total volatile basic nitrogen (TVB-N) content in duck meat based on Vis/NIR spectroscopy technique as an objective tool.