Machine vision-based tobacco leaf portion identification method, electronic device and storage medium

Abstract

The invention provides a machine vision-based tobacco leaf portion identification method, an electronic device and a storage medium. The method includes the following steps that: the standard gradingsamples of tobacco leaves are acquired, tobacco leaf sample images are collected, and all qualified tobacco leaf sample images are obtained; tobacco leaf regions where tobacco leaves are located are segmented from the qualified tobacco leaf sample images on the basis of differences between tobacco leaf colors and background colors; the mean values and standard deviations of the RGB and HSV of thetobacco leaf regions, as well as the area, length and width of the tobacco leaves are obtained; and the mean values and standard deviations of the RGB and HSV as well as the area, length and width ofthe tobacco leaves are marked as feature vectors for tobacco leaf grade identification, and a feature matrix is formed according to the feature vectors of different tobacco leaves; the feature matrixis decomposed, so that an identification model for tobacco leaf portion identification is established; and tobacco leaf portions are identified according to the identification model. With the method of the invention adopted, portion confusion in a large number of tobacco leaf portions and errors in grade identification can be avoided. The method has accurate and rapid identification capacity.

Description

technical field [0001] The invention relates to the field of tobacco technology, in particular to the field of image machine vision technology, and specifically relates to a machine vision-based tobacco leaf part recognition method, electronic equipment and a storage medium. Background technique [0002] The machine vision system can quickly acquire a large amount of tobacco leaf appearance information and automatically perform data processing. It is easy to integrate with other dimensional information in production, and has the advantage of greater benchmark scale invariance in improving the flexibility and rapid evaluation of tobacco leaf sorting. In some application fields where manual operations are frequent and artificial vision is difficult to meet the requirements, machine vision is often used to replace artificial vision. Generally, in the process of industrial grading, there are three main factors of appearance quality, "uniformity of color", "brightness of color", ...

AI Technical Summary

Problems solved by technology

[0007] In view of the shortcomings of the prior art described above, the purpose of the present invention is to provide a method for identifying parts of tobacco leaves based on machine vision, electronic equipment and storage media, to solve the problem that the parts of tobacco leaves cannot be quickly and accurately identified in the prior art

Images