Origin Identification of Poria cocos Based on Terahertz Time-Domain Spectroscopic Techniques

Abstract

Poria cocos is an important traditional Chinese medicinal material. The quality and efficacy of Poria cocos grown in different origin places are significantly different. The purpose of this study is to accurately identify the origin of Poria cocos using terahertz time-domain spectroscopy. The refractive index and the absorption coefficient spectra in the range of 0.4˜1.5 THz of Poria cocos samples from twelve provinces across China had similar change trends, with a little difference from the intensity. Combined with principal component analysis (PCA), the absorption coefficient spectrum was established to identify the origin of Poria cocos based on random forest (RF), radial basis function neural network (RBFNN) and convolutional neural network (CNN). The experimental results show that the RBFNN algorithm has the best identification effect among the 12 different origins of Poria, and the correct rate of the test set can be up to 98.2609%, which is higher than that of the RF algorithm (97.3913%) and the CNN algorithm (93.913%). Accurate identification of the origin of Poria cocos was realized with terahertz time-domain spectroscopy, providing a new idea and technical solution for the origin identification of Poria cocos.