论文部分内容阅读
目的运用近红外光谱对生鲜猪肉新鲜度进行实时评估。方法利用多通道可见近红外光谱系统,获取了猪肉表面380~1080nm波长范围内的漫反射光谱数据,采用多元散射校正(MSC)和变量标准化(SNV)的预处理方法,然后使用偏最小二乘回归建立猪肉新鲜度的预测模型,进而对猪肉新鲜度进行评价。结果采用变量标准化处理后的偏最小二乘回归模型相对比较稳定,建模效果比较好。对挥发性盐基氮(TVB-N)的验证集的相关系数达到0.91,对pH值的验证集的相关系数达到0.93。最后利用该模型对猪肉新鲜度进行评定,评定准确率达92.9%。结论实验中运用多点的测量方式提高了近红外检测的精度和稳定性,对于实时检测评估生鲜猪肉的新鲜度有很大的潜力。
Objective To assess the freshness of fresh pork by near infrared spectroscopy. Methods The multi-channel visible near-infrared spectroscopy system was used to obtain diffuse reflectance spectra in the wavelength range from 380 nm to 1080 nm on the surface of pork. The method of multivariate scatter correction (MSC) and variable normalization (SNV) Return to the establishment of a predictive model of pork freshness, and then evaluate the freshness of pork. Results The partial least squares regression model standardized by variables was relatively stable and the modeling effect was better. The correlation coefficient of the validation set for volatile basic nitrogen (TVB-N) reached 0.91, and the correlation coefficient for the validation set of pH reached 0.93. Finally, the model was used to evaluate the freshness of pork, the evaluation accuracy rate reached 92.9%. Conclusion The use of multi-point measurement in the experiment improves the accuracy and stability of near-infrared detection and has great potential for real-time detection to assess the freshness of fresh pork.