Abstract:

This study reports on the application of near-infrared (NIR) diffuse reflectance spectroscopy for rapid determination
of fat, protein and water in emulsion-type sausage. After the original spectra were calibrated and the optimum range was
selected, partial least-squares model (PLS) models for quantitative analysis of fat, protein and water were developed. The
statistical data showed that the determination coefficient (R2), root mean square error of cross validation (RMSECV), root
mean square error of prediction (RMSEP) and residual prediction deviation (RPD) for the quantitative analysis model of fat
were 0.9867, 0.556%, 1.25% and 3.98; the values of R2, RMSECV, RMSEP and RPD were 0.9723, 0.28%, 0.59% and 2.83 for
protein; and 0.9905, 0.428%, 1.15% and 3.85 for water, respectively. The results prove that the quantitative analysis models
have good precision and stability, and can be used for rapid detection of fat, protein and water in emulsion-type sausage.

Key words: emulsion-type sausage, NIR spectroscopy, rapid determination, quantitative model