期刊
FOOD CHEMISTRY
卷 424, 期 -, 页码 -出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.foodchem.2023.136411
关键词
Cow's milk; Adulteration; Chemometrics; Classification; Principle component analysis; Artificial neural networks; Support vector machine
The aim of this study was to evaluate a photoacoustic spectroscopy system with visible to short-wave near infrared (Vis-SWNIR) light sources for detecting adulterations in cow's milk. The results showed that the system was reliable and accurate, with the highest classification accuracy in detecting adulteration type and level. Further studies are suggested to generalize the findings and extend the application of the system to other liquid foods for quality evaluation.
The aim of this study is to evaluate a previousely developed photoacoustic spectroscopy system with light sources of visible to short-wave near infrared (Vis-SWNIR, 395-940 nm) for detection of adulterations in cow's milk including formalin, urea, hydrogen peroxide, starch, sodium hypochlorite, and detergent powder. The results of principal component analysis (PCA) showed a very good visual differentiation of different adulterations. The artificial neural networks (ANN) showed the highest classification accuracy (97.6 %) in detection of adulteration type and adulteration level (nearly 100 %). It can be generally concluded that the Vis-SWNIR photoacoustic spectroscopy system is a reliable and potent instrument for detecting various types of milk adulterations. Further studies are suggested with including cow's milk of different sources with probable variations in composition to generalize the findings of the present study. With the extension of the light sources to the range of long-wave NIR, the system can be applied as a diagnostic tool for quality evaluation of other liquid foods.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据