近红外光谱法检测中式滑炒猪肉的水分含量

摘要

水分含量是评价猪肉品质优劣的重要指标,与肉品的多汁性、口感和嫩度密切相关,但检测肉类水分含量的传统方法是直接干燥法,耗时耗力.实验应用傅立叶变换近红外光谱分析技术建立中式滑炒猪肉水分含量测定模型,实现快速无损检测.利用平滑结合一阶微分预处理方式并结合马氏距离、内部交叉校正均方差法(RMSECV)和学生残差剔除异常样本,建立最佳模型.结果显示:模型的相关系数(R)为0.9760,校正均方差值(RMSEC)为0.0674,且根据预测值与真实值的比较发现,预测结果正确率大于99.1%,表明该分析方法准确可靠,有望在中式烹饪猪肉水分含量检测中得到广泛应用.%Water holding capacity is an important indicator to evaluate the quality of meat products.It is closely related to the succulent,taste and tenderness of meat products.However,the traditional determination of water holding capacity of meat products is with direct drying method,which is time and labor consuming.In the study,a nondestructive and rapid detection of water holding capacity of Chinese fried pork is established by near-infrared reflectance spectroscopy (NIRS).Using smoothing combined with first-order differential preprocessing method and then mahalanobis distance,cross validation of internal crossovers (RMSECV)and studentized residual are carried out to eliminate the abnormal samples and establish the calibration model.The results show that the correlation coefficient (R)of the model is 0.9760 and the RMSEC is 0.0674.According to the comparison between the predicted value and the actual value,the correct rate of the prediction result is larger than 99.1%,indicating that the analysis method is accurate and reliable,it is expected to be widely used in the detection of pork moisture content in Chinese cooking.