真空包装冷却羊肉pH值和颜色的近红外快速无损检测

发布时间：2019-01-07 14:15

【摘要】：【目的】羊肉新鲜度理化指标传统测定中存在操作繁琐、需使用化学试剂、样品被破坏与接触等问题。研究真空包装冷却羊肉pH值和颜色的近红外快速无损检测。【方法】以新疆小尾寒羊作为研究对象,利用近红外光谱技术(NIRS),研究储藏过程中真空包装冷却羊肉理化指标的快速无损检测,针对化学指标pH和物理指标颜色亮度(L*),采用光谱预处理和特征波段筛选方法优化和简化各指标含量预测模型。【结果】pH和L*均以组合使用遗传算法(GA)与连续投影算法(SPA)筛选特征波段建立的多元线性回归模型(MLR)最佳,优于单一波段筛选方法,其预测集相关系数均为0.91,预测集均方根误差分别为0.13和1.91。【结论】近红外光谱技术具有绿色、无损、同时分析多指标的检测优势,可替代传统检测手段实现对肉品理化指标的快速无损准确检测。
[Abstract]:[objective] there are some problems in the traditional determination of the freshness of mutton, such as the complicated operation, the use of chemical reagent, the destruction and contact of the sample, etc. The pH value and color of chilled mutton in vacuum packaging were studied. [methods] using Xinjiang small tail Han sheep as the research object, the near infrared spectroscopy (NIRS),) technique was used to detect the pH value and color of mutton. The rapid nondestructive testing of physical and chemical indexes of vacuum packaged chilled mutton during storage was studied. The chemical index pH and the physical index color brightness (L *) were studied. Spectral preprocessing and feature band screening were used to optimize and simplify the prediction model of each index content. [results] pH and L * were established by combining genetic algorithm (GA) and continuous projection algorithm (SPA) to screen feature bands. The multivariate linear regression model (MLR) is the best. Compared with single band screening method, the correlation coefficient of prediction set is 0.91.The root mean square error of prediction set is 0.13 and 1.91.Conclusion Near-infrared spectroscopy has the advantages of green, nondestructive and multi-index analysis. It can replace the traditional detection method to realize fast nondestructive and accurate detection of meat physicochemical index.
【作者单位】：石河子大学机械电气工程学院;
【基金】：国家自然科学基金项目“新疆冷却羊肉储藏品质的动力学及不同光学速测技术融合研究”(31460418) 石河子大学青年创新人才培育计划项目“基于多源光学信息的羊肉新鲜度快速检测机理及系统开发研究”(CXRC201707)~~
【分类号】：O657.33;TS251.53

【相似文献】