Li Cheng-Han, Hsieh Chun-Hung, Hung Cheng-Chu, Cheng Ching-Wei
Department of Bio-Industrial Mechatronics Engineering, National Chung Hsing University, Taichung 402, Taiwan.
Department of Information Management, National Taichung University of Science and Technology, Taichung 404, Taiwan.
Foods. 2021 Feb 11;10(2):394. doi: 10.3390/foods10020394.
After completing the production of preserved eggs, traditionally, the degree of gelling is judged by allowing workers to tap the preserved eggs with their fingers and sense the resulting oscillations. The amount of oscillation is used for the quality classification. This traditional method produces varying results owing to the differences in the sensitivity of the individual workers, who are not objective. In this study, dielectric detection technology was used to classify the preserved eggs nondestructively. The impedance in the frequency range of 2-300 kHz was resolved into resistance and reactance, and was plotted on a Nyquist diagram. Next, the diagram curve was fitted in order to obtain the equivalent circuit, and the difference in the compositions of the equivalent circuits corresponding to gelled and non-gelled preserved eggs was analyzed. A preserved egg can be considered an RLC series circuit, and its decay rate is consistent with the decay rate given by mechanical vibration theory. The Nyquist diagrams for the resistance and reactance of preserved eggs clearly showed that the resistance and reactance of gelled and non-gelled eggs were quite different, and the classification of the eggs was performed using Bayesian network (BN). The results showed that a BN classifier with two variables, i.e., resistance and reactance, can be used to classify preserved eggs as gelled or non-gelled, with an accuracy of 81.0% and a kappa value of 0.62. Thus, a BN classifier based on resistance and reactance demonstrates the ability to classify the quality of preserved egg gel. This research provides a nondestructive method for the inspection of the quality of preserved egg gel, and provides a theoretical basis for the development of an automated preserved egg inspection system that can be used as the scientific basis for the determination of the quality of preserved eggs.
传统上，在完成皮蛋生产后，胶凝程度是通过让工人用手指轻敲皮蛋并感受由此产生的振荡来判断的。振荡量用于质量分级。由于个体工人敏感性存在差异且不客观，这种传统方法产生的结果各不相同。在本研究中，采用介电检测技术对皮蛋进行无损分级。将2 - 300 kHz频率范围内的阻抗分解为电阻和电抗，并绘制在奈奎斯特图上。接下来，对图曲线进行拟合以获得等效电路，并分析与胶凝和未胶凝皮蛋相对应的等效电路组成的差异。一个皮蛋可被视为一个RLC串联电路，其衰减率与机械振动理论给出的衰减率一致。皮蛋电阻和电抗的奈奎斯特图清楚地表明，胶凝和未胶凝鸡蛋的电阻和电抗有很大差异，并使用贝叶斯网络（BN）对鸡蛋进行分级。结果表明，具有电阻和电抗两个变量的BN分类器可用于将皮蛋分类为胶凝或未胶凝，准确率为81.0%，kappa值为0.62。因此，基于电阻和电抗的BN分类器展示了对皮蛋凝胶质量进行分类的能力。本研究为皮蛋凝胶质量检测提供了一种无损方法，并为开发可作为皮蛋质量判定科学依据的自动化皮蛋检测系统提供了理论基础。
