A machine vision-based method for online detection of foreign matter in shredded tobacco

Abstract

The invention discloses an on-line detection method for foreign matter after shredded tobacco based on machine vision, comprising: acquiring a color image of shredded tobacco to be detected; performing filtering and contrast enhancement processing on the color image; Extract the color space statistical features of the shredded tobacco to be detected; for the color space statistical features, use multiple preset color space classifiers for joint detection; perform statistical judgment on the detection results of each color space classifier, and make the connected domain area less than Connected domain deletion with preset area threshold; use OR operation to fuse the above judgment results to obtain pixel-level detection results; use multiple preset foreign object discriminators to identify in parallel for the above pixel-level detection results, and exclude images that do not meet the characteristics of foreign objects region; the above image data is fused to obtain the final detection result. The detection method can improve the efficiency and accuracy of the foreign matter detection of shredded tobacco.

Description

technical field [0001] The invention relates to the technical field of foreign matter detection in tobacco, in particular to an online detection method for foreign matter after shredded tobacco based on machine vision. Background technique [0002] During the cutting operation or before the operation in the cigarette processing process, there will still be residues after removal of foreign matter, and it is easy to re-mix foreign matter during the shredding process, so that the final shredded tobacco produced contains foreign matter. And once foreign matter is contained in the cigarette, it will affect the mouthfeel, and the cigarette with foreign matter enters the market like this, just upgrades to serious quality problem. At present, the common method is to use manual screening to screen the small foreign matter in shredded tobacco, which not only requires a huge workload, but also requires high labor costs, and the screening effect and efficiency are not high. Therefore,...

AI Technical Summary

Problems solved by technology

[0002] During the cutting operation or before the operation in the cigarette processing process, there will still be residues after removal of foreign matter, and it is easy to re-mix foreign matter during the cutting process, so that the final shredded tobacco contains foreign matter

Once the cigarette contains foreign matter, it will affect the taste, so that the cigarette with foreign matter enters the market, and it will be upgraded to a serious quality problem

At present, the common method is to use manual screening to screen small foreign matter in shredded tobacco, which not only has a huge workload, but also requires high labor costs, and the screening effect and efficiency are not high.

Therefore, in the process of realizing the present application, the inventors found that the efficiency and accuracy of tobacco foreign matter detection in the prior art are difficult to meet the needs of industrial production

Images