ABSTRACT
OBJECTIVE：To analysis the correlation between chrom aticity value and quality index of Atractylodis chinensis decoction piece powder stir-fired with bran ，and to determine its processing time. METHODS ：The processed samples of 16 batches of A. chinensis decoction piece stir-fired with bran （S0-S15，S0 is the raw product of A. chinensis ）were prepared ，and chromaticity values of all samples were determined ，such as lightness value （L*），yellow blue value （b*），red green value （a*）. UPLC fingerprint of sample were analyzed ，and the contents of extract and volatile oil were also determined. Pearson correlation was used to analyze the correlation between the chromaticity value and quality index （relative peak area of each chromatographic peak in UPLC fingerprint ，water-soluble extract content ，alcohol-soluble extract content and volatile oil content ）. Multivariate statistical analysis （principal component analysis ，cluster analysis ，partial least squares discriminant analysis ）was carried out with chromaticity value and quality index ，and the processing time of A. chinensis decoction piece stir-fired with bran was determined by grey correlation method. RESULTS ：In the process of bran frying ，with the extension of processing time ，L* and b* of decoction pieces powder decreased ，and a* increased first and then decreased ；relative areas of peak 1 and peak 2 increased first and then decreased，while relative areas of peak 3（5-hydroxymethyl furfural ）increased，and the areas of the other peaks decreased. The content of the extract did not change significantly with time ，and the content of the volatile oil decreased. The results of correlation analysis showed that the relative peak area of peak 2-27，alcohol-soluble extract content and volatile oil content had a certain correlation with the chromaticity value ，while the relative peak area of peak 1 and water-soluble extract content had no linear correlation with the chromaticity value. Results of multivariate statistical analysis showed that the samples were divided into mild （S0-S5），excessive （S12-S15），moderate （S6-processing time of 18-33 min）. The results of grey correlation method showed that the processing time of A. chinensis decoction piece stir-fired w ith bran should be controlled in the range of 18-24 min，and the optimal processing time was 18 min. CONCLUSIONS ：There is a correlation between chromaticity value of A. chinensis decoction piece powder stir-fired with bran and the relative peak area of 27 chromatographic peaks ，and content of extract and volatile oil. It is suggested that the processing time should be 18 min.
ABSTRACT
OBJECTIVE:To extract the feature of Achyranthes bidentata and Cyathula officinalis,and to establish image recog-nition method. METHODS:The microscopic image stitching of A. bidentata and C. officinalis was implemented by MATLAB. The color,invariant moment,stripes and the features of vascular bundle in cross section were extracted. The data was organized into da-ta matrix,and then data matrix was standardized by Zscore function;principal components were analyzed through Princomp func-tion. BP nerve network recognition mode was adopted. RESULTS:The microstructures in the micro images of the samples were kept integrated. The measured data of 27 characteristics were acquired in each group of sample. Through principal component analy-sis,the parameters of 11 main components were selected to establish BP never network. The average recognition rate of BP nerve network was 100% between 2 medicinal material relatives (n=50). CONCLUSIONS:The method can be used for micro-image auto Stitching of Chinese medicinal materials and image recognition of A. bidentata and C. officinalis.