Rapid Determination of Surimi Fat Base on Near-Infrared Spectroscopy
Objective:To apply Fourier transform near-infrared spectroscopy modeling technology to establish a rapid method for the determination of fat content in surimi.Method:The article collected a total of 215 surimi samples,including representative surimi samples of different grades and sources.After thawing and stirring,the samples were scanned on an Antaris II Fourier near-infrared spectroscopy analyzer to obtain its near-infrared spectrum and correlated with the fat data measured by the national standard method.A mathematical model for the near-infrared spectrum and surimi fat content was established through first derivative and Norris derivative filter smoothing combined with partial least squares(PLS)method and used for prediction.Result:The corrected correlation coefficient(Rc)of the quantitative model is 0.995 1,the corrected root mean square error(RMSEC)is 0.157,the predicted correlation coefficient(Rp)is 0.923 7,and the predicted root mean square error(RMSEP)is 0.111.Conclusion:The established near-infrared computational model of fat in surimi is of high accuracy,and it can be used for the rapid and efficient determination of fat content in surimi.
NETL
NSTL
万方数据
