CNN-based traditional Chinese medicinal material stir-frying degree determination method

Abstract

The invention discloses a CNN (Convolutional Neural Network)-based traditional Chinese medicinal material stir-frying degree judgment method, relates to a visual identification method, and aims to solve the problems of inaccurate judgment, difficulty in judgment of all stir-fried medicinal materials and time and labor waste caused by manual judgment of the stir-frying degree in a traditional Chinese medicinal material stir-frying process. The method comprises the following steps: S1, acquiring a plurality of images of stir-fried traditional Chinese medicinal materials; S2, labeling the image by using label open source software; s3, inputting the marked image into a cascade maskrcnn network architecture for training, and generating a regression model; S4, inputting the plurality of images to the regression model in the stir-frying process, setting the stir-frying numerical value of the traditional Chinese medicinal materials, and judging the stir-frying degree; if the quantity of the fried traditional Chinese medicinal materials is smaller than the set value, continuing frying if the frying is not completed; if the stir-fried quantity of the traditional Chinese medicinal materials is greater than or equal to a set value, reminding to stop stir-frying. Compared with manual judgment of the stir-frying degree of the traditional Chinese medicinal materials, the method is rapid and accurate, and the accuracy is 95% or above.

Description

technical field [0001] The invention relates to a visual recognition method, in particular to a recognition method for the frying degree of Chinese medicinal materials. Background technique [0002] The judgment of the existing medicinal material frying is manually judged. Experienced workers extract a part of the medicinal materials from the medicinal materials that have been fried for a certain period of time by sampling and testing, and judge whether the medicinal materials are cooked by observing the changes in the situation of the medicinal materials and comparing them with experience and standards. The degree of frying. The accuracy of manual judgment is not enough, and the sampling inspection is easy to be partial and comprehensive, and it is not easy to realize the judgment of all the fried medicinal materials. Even if it is realized, it will be quite time-consuming and laborious, and the speed is not fast enough. Contents of the invention [0003] In order to sol...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problems of inaccurate judgment and difficulty in judging all roasted medicinal materials in the process of frying traditional Chinese medicinal materials, the present invention provides a method for judging the degree of roasting of Chinese medicinal materials based on CNN

Images