| Since the authentic Fritillaria resources are scarce due to its high price and valuable medical uses,it is difficult to meet the clinical needs.Therefore,the problem of adulteration in the market is more and more serious.In the process of authenticity identification,the source of Fritillary medicine is complex,and there are many kinds of Fritillary medicine.It is difficult for ordinary people to recognize the different shapes of Fritillary medicine.Therefore,how to identify Fritillary medicine effectively and accurately becomes extremely crucial.At present,the identification of Fritillaria mainly relies on traditional trait identification,microscopic identification,physical and chemical identification,etc.The traditional identification is subjective and requires higher practical experience of operators.The microscopic identification is mainly based on the observation of the microstructure of starch grains,which requires the destruction of samples.The pretreatment of physicochemical identification is complicated and the cost is high.This paper use machine learning methods to extract the shape features of Fritillaria to achieved the purpose of authenticity identification.To this end,this paper has done the following:(1)Perform data collection and pre-processing,and establish a multi-view Fritillaria standard dataset for the first time;(2)It has use traditional machine learning algorithms to realize the classification of Fritillaria;(3)It has introduce deep learning methods to realize the identification of Fritillaria based on a single view;(4)In order to establish a more accurate classification model and better application in practical application scenarios,it has fused multi-view Fritillaria features to achieve Fritillaria classification at any view.Through the above work,this paper intends to establish a complete platform of Fritillaria identification based on machine learning,realize the automatic process for Fritillaria classification,the recognition accuracy can reach more than 84%,providing a new idea and solution for Fritillaria classification in Chinese herbal medicine industry.At the same time,the idea of this graduation project can be extended to the identification of other Chinese herbal medicine,which lays a foundation for the realization of batch automated identification of Chinese herbal medicine industry. |
