LIANG Li-jin, LU Wen-biao, WANG Feng-mei. Segmentation and Characterization of Saposhnikoviae Radix Micrograph Based on Edge Detection[J]. Chinese journal of experimental traditional medical formulae, 2018, 24(6): 37-41.
LIANG Li-jin, LU Wen-biao, WANG Feng-mei. Segmentation and Characterization of Saposhnikoviae Radix Micrograph Based on Edge Detection[J]. Chinese journal of experimental traditional medical formulae, 2018, 24(6): 37-41. DOI: 10.13422/j.cnki.syfjx.20180608.
Objective: To realize the quantitative characterization of varied morphological structures in Saposhnikoviae Radix by programming
in order to provide technical support for the quantitative description of the microscopic identification characteristics of the Chinese herbal medicines.Method: According to the characteristics of the main structures in micrograph of the materials
digital image processing technology and MATLAB R2013a software programming were used for taking and encoding cross sectional micrographs of the materials; registration and stitching for each gray image were completed by template matching; edge detection was conducted for mosaic images using Canny operator; binarization and segmentation were made for microscopic characteristics to analyze the size
shape and number by basic morphological operation; distribution characterization was made for oil tubes and vessels using by hierarchical clustering; and discriminant analysis was performed for the data of the samples from four origins.Result: The main microscopic characteristics in the transverse section images of the materials
such as phellem
phloem
xylem
oil tubes and vessels
were respectively segmented into sub images; 20 parameters were extracted from the sub images
and 12 parameters of them were screened out by stepwise discriminant analysis. Through the discriminant analysis of the morphological parameters
the total accuracy of the sample discrimination was 91.4%.Conclusion: The method can accurately describe the fine differences of the transverse section tissues of the samples
which has certain reference significance for the microscopic identification of the medicinal materials from different origins.