论文部分内容阅读
挥发性盐基氮(TVB-N)含量是评价猪肉新鲜度的重要指标。尝试融合光谱和成像技术检测猪肉中TVB-N含量。实验以不同新鲜度的猪肉样本为研究对象,同时采集近红外光谱数据和图像数据,并对其分别进行特征提取和主成分分析,利用反向传播神经网络构建猪肉TVB-N的定量预测模型。实验结果表明,融合模型要优于单一技术模型,模型交互验证均方根误差(RMSECV)为1.2975,对独立样本预测时相关系数达到0.957。研究表明基于光谱和成像融合技术检测猪肉中TVB-N含量是可行的,检测结果的准确性和稳定性较单一技术有所提高。
The content of volatile basic nitrogen (TVB-N) is an important index to evaluate pork freshness. Try Fusion Spectroscopy and Imaging Techniques to Detect TVB-N Content in Pork. In this experiment, different freshness pork samples were taken as research object, near-infrared spectroscopy data and image data were collected at the same time, and their feature extraction and principal component analysis were respectively carried out. The quantitative prediction model of pork TVB-N was established by backpropagation neural network. The experimental results show that the fusion model is superior to the single technical model, the root mean square error of validation (RMSECV) of the model is 1.2975, and the correlation coefficient of independent samples is 0.957. The results show that it is feasible to detect the content of TVB-N in pork meat based on spectral and imaging fusion technology. The accuracy and stability of the detection results are better than that of single technology.