Identification of the Regions of Rosemary by Ultraviolet and Visible Spectrophotometer Combined with Artificial Neural Network
To identify the origin of rosemary,a rapid identification method based on ultraviolet spectroscopy was developed.The UV-VIS spectra of 200-700 nm of rosemary from different origin were collected,and the original spectral data,first derivation and second derivation,and the data after wavelet denoising,first derivation and second derivation after wavelet denoising were carried out principal component analysis to compare the origin identification effect.Artificial neural network was used to predict the origin of rosemary.The results indicated the first two principal components of the first derivative data after wavelet denoising had the best effect on distinguishing the rosemary samples from different origins.The adjusted cosine similarity of the first derivative data of rosemary samples from different origins after wavelet denoising was between 0.823 and 0.999.Artificial neural network was used to model and predict the first order derivative data of rosemary from different origin after spectral wavelet denoising.The identification rate of rosemary was 100％.The UV-vis spectrum combined with wavelet noise reduction and neural network could quickly identify the origin of rosemary.
NETL
NSTL
万方数据
