Fritillaria cirrhosa apparent feature extraction method

Abstract

The invention relates to the technical field of image processing and feature extraction of traditional Chinese medicines. The method comprises a tendril-leaved fritillary bulb image acquisition equipment manufacturing and image acquisition stage and a tendril-leaved fritillary bulb feature extraction stage. Wherein the Fritillaria cirrhosa image acquisition equipment manufacturing and image acquisition stage comprises the steps of equipment 3D model design, entity equipment manufacturing and Fritillaria cirrhosa image acquisition. The tendril-leaved fritillary bulb feature extraction stage comprises the step of extracting the area, the texture definition, the small petal proportion, the shape, the bottom concavity and convexity and the color of tendril-leaved fritillary bulbs; according to the method, the bulbus fritillariae cirrhosae image acquisition problem under the unified environment condition is solved, so that the bulbus fritillariae cirrhosae features are accurately extracted, the bulbus fritillariae cirrhosae can be classified according to the extracted features, the interpretability is high, the practicability is high, and the feature extraction method in the field of traditional Chinese medicine is perfected.

Description

technical field [0001] The invention belongs to the field of image processing, and relates to a method for extracting apparent features of Chuanbei. Background technique [0002] In the traditional Chinese medicine industry, the currently used identification technologies mainly include manual identification and DNA identification technology. Among them, manual identification technology has low identification efficiency and large errors; DNA identification technology has high accuracy but extremely low efficiency. For traditional Chinese medicines, especially the number of pseudo-mixtures of Chuanbei, it is necessary to invent an identification method combined with the field of artificial intelligence to identify Chuanbei quickly and efficiently. [0003] For DNA identification technology, Chen Shilin's team established a DNA barcode identification system for Chinese medicinal materials. [0004] According to the "Research Status, Problems and Countermeasures of the Identifi...

AI Technical Summary

Problems solved by technology

This feature extraction algorithm uses the images of the four faces of Fritillaria, which increases the complexity of image collection, but draws the conclusion that the front view of Chuanbei is the most prominent feature of Chuanbei

However, researchers focus on academic research on the identification of Fritillaria, without practical application, the extracted features are relatively abstract, not very interpretable, and other important features required for the classification of Fritillaria have not been extracted

[0006] Based on the above analysis content, it can be seen that various Chuanbei identification techniques need to be improved, and feature extraction techniques need to be improved.

Images