Tobacco shred structure detection method based on image processing

Abstract

A tobacco shred structure detection method based on image processing comprises the following steps: mixing cut tobaccos in the three length intervals according to different mass ratios to obtain multiple groups of cut tobaccos, recording the proportion value, respectively flattening each component tobacco shred, collecting the image of each component tobacco shred, preprocessing the images, obtaining a characteristic image of each component image, calculating the texture characteristic data volume, establishing a digital model of the image texture characteristic data volume and the tobacco shred with different lengths in percentage by mass, and calculating a parameter matrix K and a parameter matrix B according to a plurality of groups of data; and collecting the image of a to-be-detectedmixed tobacco shred, preprocessing the to-be-detected mixed tobacco shred image, obtaining a texture feature image of the to-be-detected mixed tobacco shred image, calculating a texture feature data size, and calculating the mass percentages of tobacco shreds with different lengths in the to-be-detected mixed tobacco shred image in combination with the established digital model. When the tobacco shred structure detection method is used, automatic online detection of the tobacco shred structure can be achieved, and the detection efficiency is high.

Description

technical field [0001] The invention relates to the technical field of tobacco structure detection methods, in particular to a tobacco structure detection method based on image processing. Background technique [0002] In recent years, with the continuous updating of cigarette equipment and the emergence of high-speed cigarette machines, different shredded tobacco structures have a significant impact on the amount of doffing at the end of cigarettes. Unreasonable structure of shredded tobacco will cause the burning cone of cigarettes to fall off easily. The difference in shredded tobacco structure not only affects the physical quality and smoking quality of the product, but may also affect the coiling process. The current method for determining the structure of shredded tobacco is to pass the shredded tobacco sample through a shredded tobacco vibrating sorting sieve to separate shredded tobacco of different lengths, and the result is expressed by the ratio of the cumulative ...

AI Technical Summary

Problems solved by technology

[0004] (1) From sampling to weighing, it mainly depends on manual work, which is labor-intensive, has a long sampling cycle, and is more affected by human factors, resulting in poor evaluation objectivity;

[0005] (2) Data can only be entered manually, which is not compatible with the modern production and management of tobacco companies;

[0006] (3) Since the test is carried out by sampling, there is no real-time performance. If the result shows that it is unqualified, a lot of shredded tobacco has entered the next production line, which will affect the quality and production efficiency of cigarette production.

Images