Application Of Terahertz Time-Domain Spectroscopy Combined With Support Vector Machine To Determine Tea And Pesticide Samples

The transmission-type terahertz spectrometer can rapidly identify samples in a nondestructive and non-contact way. This method can also identify the important components in these samples. To test the application of the Terahertz Time-Domain Spectrometer (THz-TDS), our work focused on the identifying isomers of catechin (an important compound in tea) and the difference among green, black, and white tea. We detected the properties of pesticide residues such as thiamethoxam. Thiamethoxam with similar structure was added into the tea samples to test the resolution of the system. In this work, Support Vector Machine (SVM) and K-means clustering algorithms were employed to identify green, white, and black tea. The terahertz plus identification spectra of these structures were obtained from the experiment. The results demonstrate that THz-TDS has strong physical recognition abilities and supports the theory and application of terahertz spectrum and terahertz response.