Abstract :
With rapidly analysis, no pollution, no damage, simple operation, low analysis cost, environmental protection and many other advantages, the near infrared spectroscopy (NIR) analysis has made breakthrough progress in the Chinese medicine field. In this paper, the near infrared spectrometry of extract of two kinds of astragalus is determined. Wavelet transform is used to compress the spectral variables, and the quantitative analysis models are carried on using artificial neural network technology in order to analyze the astragaloside content of extract of two kinds of astragalus. The simulation results show that, the prediction decision coefficient(R2) is 0.9863, the average relative error is 0.0354, the root mean square error of Cross-Validation(RMSECV) is 0.0258 in the astragalus extract samples (the ratio of material to liquid 1:2), and the predictive decision coefficient is 0.9798, the average relative error is 0.0425, and the root mean square error of Cross-Validation is 0.0301 in the astragalus extract samples (the ratio of material to liquid 1:5). The evaluation model can meet the need of practical application, and provide technical support for quantitative analysis to extract of astragalus and analysis of near infrared spectroscopy in traditional Chinese medicinal materials.
Keywords :
botany; data compression; infrared spectroscopy; mean square error methods; neural nets; spectroscopy computing; wavelet transforms; Chinese medicine field; NIR analysis; RMSECV; artificial neural network technology; astragaloside near infrared spectroscopy quantitative analysis; prediction decision coefficient; root mean square error cross-validation; sample extraction; wavelet transform; Analytical models; Artificial neural networks; Liquids; Spectroscopy; Wavelet transforms; artificial neural network; extract of astragalus; near infrared spectroscopy; quantitative analysis; wavelet transform;
