论文部分内容阅读
目的：建立中药组织显微图像定量分析与鉴定方法。方法：以改进的Ｍａｓ－Ⅱ定量图像分析系统测试各种郁金类中药组织显微图像的体视学形态学参数，并结合数值分类手段，定性定量地刻画郁金类中药显微鉴定特征及评判标准。结果：当分类阀值Ｌ１＝８．２５时，基本上可将不同品种的郁金类中药区分开来；内皮层细胞平均最大直径等１０个因子对郁金类中药显微鉴别影响程度最大。结论：基于组织形态体视学和数值分类学的中药显微定量图像模式识别技术科学可行，尤适于近缘品种及易混淆中药的鉴别。
Objective: To establish a quantitative analysis and identification method for microscopic images of traditional Chinese medicine. Methods: The improved Mas-II quantitative image analysis system was used to test the stereoscopic morphological parameters of the micrographs of various turmeric tissue of traditional Chinese medicine, and combined with numerical classification methods to qualitatively and quantitatively characterize the microscopic identification characteristics of turmerics. Judging criteria. Results: When the classification threshold L1 was 8.25, different varieties of turmeric herbs could be distinguished from each other. The average maximum diameter of endothelial cells and other 10 factors had the greatest influence on the microscopic identification of turmeric herbs. Conclusion: The traditional Chinese medicine microscopic quantitative image pattern recognition technology based on tissue morphology stereology and numerical taxonomy is scientific and feasible, especially suitable for the identification of closely related species and easily confused Chinese medicine.