近红外光谱技术预测猪肉的pH、嫩度和蒸煮损失

以冷却猪肉为研究对象，评价近红外光谱（NIR）技术用于肉类物理特性预测的可行性以及不同的光谱处理方法和建模方法对预测准确性的影响。试样取自排酸24h的同一批猪胴体的小里脊肉，采集4000—10000cm-1的光谱。经外部验证的偏最小二乘（PLS）模型在预测pH时表现出良好的相关性（Rc^2=0．88，Rp^2=0．80，SEC=0．08，SEP=0．084），嫩度与蒸煮损失模型的相关性分别是Rc^2=0．50和0．57，R；=0．34和0．50。在各种光谱预处理方法中，平滑处理结合多元散射校正（MSC）或标准正态变量变换（SNV）的效果最好。

Predicting pH values,tenderness and cooking loss of chilled pork based on near infrared spectroscopy

The feasibility by using near infrared spectroscopy （NIR） to predict physical attributes of chilled pork was focused on, and the effects of different methods of preprocessing spectra and modeling on the accuracy of the established models were studied. The samples were selected from the undercut of one batch carcasses after 24 h post-mortem,the diffuse reflectance spectrams with 4 000-10 000 cm-1 scan- ning information were collected for modeling. The results showed that the mathematical model for predic- ting pH values of chilled pork by PLS after validation test was proved and it had higher correlation param- eters （Rc^2=0. 88 ,Rp^2 =0. 80 ,SEC =0. 08 ,SEP =0. ones （Rc2 = 0. 50 and 0.57, Rp^2 = 0. 34 and 0.50）. 08 ）, while for tenderness and cooking loss with lower Moreover, the study also indicated that the method for smoothing spectra combined with multiplicative scatter correction （MSC） or standard normal variable transformation （SNV） was better than other methods of pre-treatment spectra.